Dehumanization in Everyday Politics A study of discursive dehumanization of beggars on social media Author: Joanna Hellström Uppsala University, Fall 2018 Department of Government Master thesis, 30 hp Supervisor: Daniel Nohrstedt
Dehumanization in Everyday Politics
A study of discursive dehumanization of beggars on social media Author: Joanna Hellström Uppsala University, Fall 2018 Department of Government Master thesis, 30 hp Supervisor: Daniel Nohrstedt
2
Abstract Social science scholars mainly regard dehumanization as a phenomenon of conflict and
war. Concurrently, dehumanizing attitudes and behavior in democratic, non-conflict
settings is a significant field within social psychology. Given the rise of right-wing
populism and populist rhetoric, there is reason to believe that dehumanization has
nestled its way into political discourse. With this background, the current study has
investigated whether dehumanizing attitudes allegedly held by citizens are also
expressed in their political arguments. Dehumanization can lead to support for
aggression, discrimination, and violence even in democratic societies. Therefore, it is
essential for political scientists to acknowledge its existence, as dehumanization can
have a severe impact on equality and the defense of human rights. This paper is a cross-
cutting study, which bridges the gap between political science and social psychology in
the study of dehumanization by answering the question: Whether and how
dehumanization is used in the debate on street begging in everyday politics?
Discursive dehumanization on social media was mapped with the help of a novel
analytical tool for content analysis. Social psychology has shown that people hold
dehumanizing attitudes, and this study has shown that people are also willing to express
these attitudes. This study finds that dehumanization is used in the debate on street
begging to a significant degree. Dehumanization is mainly used in negative depictions
of beggars, which undermine their moral capacity, civility, and refinement. The results
of this study provide a foundation for studying both discursive dehumanization and
dehumanization in political science.
Keywords: Dehumanization, political communication, discourse, right-wing populism, social media, political participation
3
Table of Contents
1INTRODUCTION 41.2PURPOSEANDRESEARCHQUESTION 71.3STRUCTUREOFTHETHESIS 7
2THEORY 82.1ANTI-IMMIGRATIONRHETORICANDATTITUDES 82.1.1PARTYRHETORIC 92.1.2SENSEOFTHREAT 112.2EVERYDAYPOLITICS 132.3DEHUMANIZATION–ANOVERVIEW 172.3.1INTRODUCTIONTOTHEFIELDOFDEHUMANIZATION 172.3.2THECONCEPTOFDEHUMANIZATION 192.3.3DEFININGDEHUMANIZATION 222.3.4RELATIONSHIPTOOTHERCONCEPTS 24
3METHODOLOGY 263.1DESIGN 263.2METHOD 263.2.1CASESELECTION 273.3MATERIAL 283.3.1DATACOLLECTION 293.3.2ETHICALCONSIDERATIONS 313.4DATAANALYSIS 313.4.1OPERATIONALIZATION 313.2.2OPERATIONALINDICATORS 333.2.3WHATISNOTDEHUMANIZATION? 34
4RESULTSANDDISCUSSION 354.1RESULTS 354.1.1GENERALRESULTS 354.1.2PARTYPAGESPECIFICRESULTS 364.1.3ELEMENTSOFDEHUMANIZATION 384.2DISCUSSION 424.2.1.MAKINGSENSEOFTHE22PERCENT 424.2.2DEHUMANIZATIONANDPARTYSUPPORT 434.2.3THERELATIVEPREVALENCEOFMORALDEHUMANIZATION 444.2.4THEAPPLICABILITYOFTHERESULTS 45
5CONCLUSION 485.1WHETHERDEHUMANIZATIONISUSED 485.2HOWDEHUMANIZATIONISUSED 495.3THEFUTUREOFDEHUMANIZATIONINPOLITICALSCIENCE 50
6LISTOFREFERENCES 52
7APPENDIX 62APPENDIX1:INDICATORS 62APPENDIX2:CODESHEET 65APPENDIX3:APPLYINGTHEANALYTICALTOOL 66
4
1 Introduction Dehumanization is the act of perceiving someone as less than human (Haslam and
Stratemeyer 2016). It is seen as an explanation for why soldiers, correctional officers,
and even ordinary citizens can commit horrendous crimes against other human beings.
Dehumanizing metaphors were used as propaganda tools during the Holocaust and the
genocide in Rwanda, where the extermination of Jews and Tutsis were preceded by
depictions of them as vermin and pests (Livingstone Smith 2011). Despite this,
dehumanization under normal political circumstances in democratic societies has been
somewhat forgotten by social and political science. These everyday dehumanizing
processes have, however, emerged as a central field of research within social
psychology. At the same time, there is a trend towards right-wing populism and a
hardening political climate across Europe (Wodak 2013; Ingelhart & Norris 2016). This
begs the question of whether it has become accepted for citizens to express
dehumanizing views in the political debate. The sphere in which citizens express their
political opinions, outside formal and traditional political forums is called everyday
politics (Highfield 2016). This sphere is a place where a fifth of the Swedish population
expresses their political opinions (Internetstiftelsen 2018).
Political Science and neighboring fields have mainly studied dehumanization as a
mechanism of conflict and war, yet social psychology finds that even ordinary people in
peaceful, democratic societies hold dehumanizing views. This is called everyday
dehumanization (Livingstone Smith 2011; Hagan & Rymond-Richmond 2008; Oliver
2015; Savage 2013; Roscoe 2007; Luft 2015; Haslam & Loughnan 2014). Individuals
who hold dehumanizing views have been known to take action against their victim,
support acts of violence, and to stay passive when dehumanized groups are in need
(Lindén et al. 2015; Bruneau, Kteily & Laustsen 2018; Andrighetto 2015; Cuddy 2007).
Dehumanization can have severe consequences even in democratic countries. It is
regarded as the most severe form of intergroup prejudice (Bruneau, Kteily & Laustsen
2018) and can lead to support for aggressive retaliatory policies even against vulnerable
groups (Maoz & McCauley 2008). Dehumanization is often used as a justification for
violence and differential treatment, even in democratic, peaceful societies (Volpato &
Andrighetto 2015). With this background, it is critical to understand whether
dehumanization is a part of the democratic political debate. Using dehumanizing
5
narratives about others can threaten democratic principles of equality, and lead to the
acceptance of human rights violations that occur right in front of our eyes. From a
theoretical perspective, research on democracy and political participation can learn from
this study to increase the understanding of the complex relationship between social
media and democracy. At the same time as social media are accessible and resource
effective means of participation, which provide possibilities for collective action;
scholars need to understand that extremism and incitement is also a part of this image.
In between the dark and the light side of social media, there is also a grey area, which is
explored in this paper.
The rise of right-wing populism, toughening political climate, and increasingly
polarized political debate are all a part of the recent political development in Western
societies (Wodak 2013; Ingelhart & Norris 2016). Political science has offered different
explanations for why people support restrictive immigration policies, and which
strategies politicians use to make them do so. The hegemonic theories of economic
inequality as the sole factor have been questioned, and the importance of psychological
factors have gained increased recognition among scholars (Norris & Ingelhart 2016;
Lucassen & Lubbers 2012). Political rhetoric and perception of threat from immigration
have been highlighted as two of the contributing factors (Mudde 2007; Rydgren 2007;
Wirz et al. 2018; Lucassen & Lubbers 2012; Ivarsflaten 2008). Concurrently, research
in social psychology shows that exposure to dehumanizing portrayals of immigrants
leads to negative attitudes towards immigration in general (Utych 2018; Goméz-
Martínez & de la Villa Moral- Jiménez 2018). In addition, several studies find that
dehumanization mediates the relationship between threat perception and negative
attitudes towards immigration (Maoz & MacCauley 2008; Louis, Esses & Lalonde
2013; Prati et al. 2016).
Dehumanizing imagery is theoretically compatible with populist rhetorical form and
content. Since dehumanization is based on making generalizations of other groups,
especially those who have been traditionally discriminated against (Leyens et al. 2000),
it can also be a convenient tool for radical right-wing parties. It is therefore important to
study political argumentation through the lens of dehumanization: to investigate
whether the rise of right-wing populism brought discursive dehumanization
(dehumanization through the use of language) into everyday politics. This study can
6
thus provide a foundation for cross-cutting studies between social psychology and
political science, where both discourses can learn from each other when it comes to
studying anti-immigrant attitudes.
Research on everyday dehumanization studies emotions, attitudes, and effects of
dehumanization among citizens in non-conflict settings, predominantly western
democracies. The expressions of dehumanization, which are studied in this context, are
often implicit and partially dehumanizing (Haslam & Loughnan 2014; Volpato &
Andrighetto 2015). An emerging field in political science and political participation is
everyday politics: the political sphere of ordinary citizens (Highfield 2016). Social
psychology has found that dehumanization exists in the form of perceptions and
attitudes in the sphere of the everyday. While social psychology can provide insight into
the neural and emotional mechanisms of everyday dehumanization, political science can
provide the political outlook into dehumanization in everyday politics that engage
citizens. Therefore, it is reasonable to inquire whether citizens in their everyday
political activities also express dehumanizing views. In the current study,
dehumanization is analyzed by using a novel tool built on previous research in social
psychology. The extent and use of dehumanization in the debate on begging is mapped
through a quantitative content analysis of social media comments. The advantage of
using this novel approach to dehumanization is that it enables the study of subtle forms
of dehumanization. Furthermore, it aids in the process of bringing order to the
conceptional quagmire of social science studies on dehumanization. I will return to this
issue in chapter 2.
By studying discursive dehumanization in everyday politics the contribution that this
study makes is threefold: 1) To provide a foundation for studying dehumanization in
political science by connecting it to right-wing populism, 2) as well as creating an
analytical instrument for studying subtle forms of discursive dehumanization. By doing
this, the study also 3) complements existing studies on everyday dehumanization by
exposing that the dehumanizing views that ordinary people hold are also reflected in
their political views and the arguments that they put forward.
7
1. 2 Purpose and Research Question The ambition of this cross-cutting study is to bridge the gap between right-wing
populism and dehumanization, political science and social psychology. This is going to
be achieved by applying a theoretical framework from social psychology to everyday
politics. By doing this, the study will also complement existing studies within the field
of dehumanization. The purpose of this thesis is thus to examine manifestations of
dehumanization in everyday politics with the question:
Whether and how dehumanization is used in the debate on street begging in everyday
politics?
Since there is no data on the prevalence of dehumanizing arguments in the political
debate, the study sets out to answer whether dehumanization is used in social media
discussions for and against begging. Given that this is a descriptive study, the question
of how dehumanization is used is answered by examining which elements of
dehumanization that are expressed in the debate, and their frequencies.
1. 3 Structure of the Thesis This paper consists of 5 chapters. Chapter 2 introduces the theoretical background and
gives the foundation for studying dehumanization as a phenomenon and within political
science. The third chapter gives a detailed description of how the study has been carried
out regarding method and material selection as well as operationalization and analytical
framework. Chapter 4 gives an outline of the results and discusses their implications in
the light of previous research. Chapter 5 concludes the study results. Appendix 1-3
shows the analytical tool, the code sheet and examples of how they were applied.
8
2 Theory
The theory chapter is divided into three parts: anti-immigration rhetoric and attitudes,
everyday politics, and dehumanization. The first part places dehumanization within a
political science discourse and proposes possible connections between dehumanization
and hostile attitudes towards immigration. The second part of the theory chapter deals
with previous research on everyday politics: political participation and attitudes online.
The final part consists of a brief outline of dehumanization as a research field, concept
and definitions, and relationship to other concepts.
2.1 Anti-immigration rhetoric and attitudes
The purpose of this paper is to place dehumanization within a general discussion in
political science and connect it to state of the art discussions on migration issues. As
previously explained, scholars within social psychology have found evidence of a
correlation between dehumanization and actual anti-immigrant sentiments. Modern
research on anti-immigration rhetoric and attitudes is mostly conducted under the larger
umbrella of right-wing populism. This begs the question of how dehumanization as a
factor fits into the debate on immigration policy and the rise of radical populist right-
wing parties. This section is therefore devoted to a brief sketch of right-wing populist
rhetoric connected to dehumanization. Thereafter the relationship between threat
perception and dehumanization is outlined.
Negative attitudes towards immigration are what distinguish the electorate of populist
radical right parties from that of other parties (Norris 2005). Scholars within political
science have proposed several explanatory factors for why people vote for populist
right-wing parties or parties on the extreme right. These factors can be divided into
micro, meso and macro levels of analysis where we have both demand and supply side
variables (Mudde 2007). Among micro-level explanations, nativism (the preference for
native people before immigrants) is a crucial factor behind populist radical right-wing
voting (Ivarsflaten 2008; Mudde 2007; Norris 2005; Rydgren 2008). Nativism bares
resemblance to dehumanization, as the latter is a severe form of devaluation and
essentialization of other groups where they are excluded from the realm of humanity
9
(Leyens et al. 2000). By extension, dehumanization strengthens the feeling of group
identity and is often used as a way of legitimizing differential treatment (Volpato &
Andrighetto 2015).
2.1.1 Party rhetoric
Beyond ideology, political party propaganda and rhetoric has been proposed as an
important factor behind the attraction of populist radical right parties (Mudde 2007;
Rydgren 2007). Rhetoric, in this case, can be divided into message content and message
form respectively (Engesser et al. 2017). Message content denotes what is being
presented and message form means how the message is presented, in terms of style and
technique. Populist radical right actors employ a combination of radical right and anti-
immigration messages with a populist rhetorical toolbox (Wirz et al. 2018). Rhetorical
content of right wing-populist actors contain elements of exclusion of certain societal
groups, while parties on the more radical end use harsh anti-immigration
communication (Mudde 2007; Wirz et al. 2018).
Dehumanizing metaphors are a documented part of anti-immigration content, even in
present day. On a general level, dehumanization can be utilized to justify all kinds of
action from inequalities to violence. Individuals might seek to justify and protect the
status quo and their position of privilege by adhering to an idea that refugees or "others"
are undeserving of the same treatment. It can thus, be used as a way to relieve feelings
of guilt that might arise from the mistreatment of someone (Volpato & Andrighetto
2015; Esses, Medianu & Lawson 2013). The imagery can play into a politics of fear,
which can be utilized politically in order to mobilize support for restrictive immigration
policies (Bleiker et al. 2013).
Metaphors depicting people as vermin and pests is something most people associate
with the infamous depictions of Jewish people as rats during the Holocaust, or Tutsis as
cockroaches during the genocide in Rwanda (Livingstone Smith 2011). Still, this form
of animalistic dehumanization exists even today when the media, politicians, and
opinion makers argue for restricted immigration. For example, refugees have been
depicted as vermin and carriers of disease in Canadian media. The claim is that if
refugees are allowed into the country, they will spread infectious disease and
10
contaminate the environment (Esses, Medianu & Lawson 2013). In 2015 a columnist in
the Sun depicted refugees in Calais as a plague of cockroaches, and former British
Prime Minister David Cameron indirectly referred to migrants coming to the UK as
insects by calling them "a swarm of people coming across the Mediterranean…"
(Anderson 2017: 13). In these xenophobic statements, dehumanizing metaphors are
used as a way of expressing that migrants are an alleged threat as they come in large
numbers and are implied to be a contagion.
Although there are many similarities, it is worth noting that not all dehumanizing
statements take a stance against immigration, and not all arguments against immigration
are dehumanizing. For instance, “taking care of group x is not our responsibility” is an
example of an argument against receiving refugees or economic migrants. Yet, it is not
an example of dehumanization.
It is predominantly the ideology of the populist radical right that we associate with
dehumanization. However, a populist rhetoric style can lend itself to dehumanizing
metaphors. The idea of the people versus the elite is often expressed through the use of
simple, ordinary language that can appeal to the man in the street as well as polarizing
rhetoric where someone is either friend or foe (Bos et al. 2013; Engesser et al. 2017).
Immigrants and asylum seekers are often a target when populists assign blame (Jagers
& Walgrave 2007; Engesser et al. 2017). One example is the sentiment refugees are
bogus, which is an example of moral dehumanization of refugees (Esses, Medianu &
Lawson 2013). The statement is a simplified way of expressing the opinion that
refugees are not really in need, but are immoral people who would not stop at any
measure to get into a country (Esses et al. 2008). Consequently, it is also an example of
populist rhetorical form as it paints a clear picture of a foe by using very simple
language. The populist rhetorical form should not be discarded as mere rhetoric. It is
important to acknowledge the possible impact of the rhetorical tools themselves. For
example, an experimental study by Bos et al. (2013) found that populist style had a
positive impact on their perceived legitimacy among lower educated and politically
cynical voters.
11
2.1.2 Sense of Threat An increasing amount of scholars are beginning to question the economic inequality
theory as the sole reason for people to support populist parties (Norris & Ingelhart 2016;
Lucassen & Lubbers 2012). One pervasive theoretical macro-level explanation in the
literature is a perceived ethnic threat from immigration. As concluded by Mudde
(2007), these studies provide mixed results. Many of these studies investigate the
relationship between the level of immigration or refugee intake and support for populist
radical right parties. Consequently, this is at best a proxy for a perceived threat, as we
cannot equate immigration figures with perceptions. It is therefore critical to distinguish
realistic threats, that is, actual competition over resources such as jobs and welfare from
symbolic threats, i.e. the anticipation or fear that a difference in values, norms and
morality may result in adverse outcomes (Stephan et al. 1998). More recent studies have
used surveys to directly measure perceived economic or cultural threats on micro-level,
where the latter has been highlighted as the most significant (Lucassen & Lubbers
2012).
The claim that immigration is a threat to norms, culture or social cohesion has been
highlighted as a crucial way for populist right-wing parties to mobilize support
(Ivarsflaten 2008). The experience of losing power and influence in a society makes
people especially susceptible to populist ideas: "the rise of populist parties reflects,
above all, a reaction against a wide range of rapid cultural changes that seem to be
eroding the basic values and customs of Western societies” (Norris & Ingelhart 2016).
A recent study on right-wing populist voting in Sweden lends support for this
conclusion. Among voters for the Sweden Democrats, 82-85% agree with a statement
that immigration threatens Swedish culture by weakening it, and 96-98% believe that
immigration leads to an increase in criminality. By contrast the 49 respectively 80
percent of voters for the Swedish conservative party Moderaterna shared this view. As
for the Social Democrats, only 22% agreed that immigrants posed a threat to Swedish
culture, and 43% thinks immigration increases criminality (Jylhä, Rydgren & Strimling
2018). In correspondence with earlier studies, the sense of economic threat was lower
compared to cultural threat, on 31-37% (Jylhä, Rydgren & Strimling 2018; Lucassen &
Lubbers 2012).
12
The direction of the relationship between threat perception and support for right-wing
populist parties is yet to be established. Previous studies indicate that the relationship is
bidirectional, where right-wing populist parties influence threat perception, and threat
perception, in turn, leads to support for right-wing populist parties (Berning & Schlueter
2015). It is thus possible that harsh anti-immigration rhetoric contributes to the
perception of immigrants as threatening among party followers, as well as the public
(Berning & Schlueter 2015; Rydgren 2007; Louis, Esses & Lalonde 2013). Negative
stereotypes and intergroup anxiety can shape the attitudes of citizens, especially those
with lower education levels (Matthes and Schmuck 2017; Stephan et al. 1998; Wirz et
al. 2018).
On an empirical level, sense of threat and dehumanization are two distinct
psychological phenomena even though they tend to appear together (Haslam 2006;
Maoz & MacCauley 2008; Louis, Esses & Lalonde 2013). Results of previous
experiments in social psychology show that a decreased perception of threat is
correlated with decreased dehumanization (Prati et al. 2016). In these studies, both
symbolic and realistic threats, as well as feelings of worry and intergroup anxiety are
measured (Prati 2016; Stephan & Stephan 2000). The proposed correlation between the
two is that perceived threat is an antecedent to dehumanization, and several studies have
results to support that conclusion (Maoz & MacCauley 2008; Louis, Esses & Lalonde
2013; Prati et al. 2016). More specifically, dehumanization has been established as a
mediator between sense of threat and hostile attitudes towards immigration (Louis,
Esses & Lalonde 2013).
In conclusion, the relationship between dehumanization and anti-immigration attitudes
can work in many ways. Dehumanization can play a rhetorical role in the expression of
populist and anti-immigration opinions and politicians have been known to occasionally
use dehumanizing language to convey political messages. This suggests that some of
the people who hold negative attitudes toward immigration in general and EU migrants
in particular, also express dehumanizing arguments in the debate. Therefore it is likely
that dehumanization is used to some extent in the debate, and these attitudes are likely
to be held by people who argue for a ban against street begging. In addition,
dehumanization itself can be a facilitating mechanism between threat perception and
hostile attitudes toward immigration, as some studies find. At the same time, there is a
13
correlation between perceiving immigration as a cultural threat, and having negative
attitudes towards immigration. If dehumanization mediates the relationship between
sense of threat and hostile attitudes towards immigrant, we are likely to find
dehumanization on all party pages, as party supporters across the board perceive
immigration as a threat to some extent (Jylhä, Rydgren & Strimling 2018).
2.2 Everyday politics
In this paper everyday politics is a form of political participation. Rather than politics by
politicians, everyday politics is politics by the people. This concept requires an
appreciation of politics as more than voting, protesting or political speeches and
reforms. In this day and age, everyday politics takes place on various social media
platforms (Highfield 2016). Therefore, this chapter deals with political participation and
activities online, including empirical data from Sweden to better interpret the results of
this study.
Most notably political participation on social media has gained the public’s attention
through online mobilization and collective action during the Arab spring. Whether
social media can bring about or enhance democracy, transparency, and accountability is
a disputed topic and the initial positivity around the power of Social Media has partly
faded and turned into skepticism (Bellin 2012; Halpern & Gibbs 2013; Harlow &
Johnson 2011). Some studies show a positive correlation between social media activity
and support for civil liberties and democratic norms (Swigger 2013; Placek 2017). At
the same time, Facebook is being used as a tool to spread hatred and violence. United
Nations investigator to Myanmar established that Facebook had played a role in inciting
violence against the Rohingya people: “It[social media] has ... substantively contributed
to the level of acrimony and dissension and conflict, if you will, within the public. Hate
speech is certainly of course a part of that.” (Miles 2018).
Notwithstanding this discussion, social media platforms are important forums for
citizens to express, debate and share their political opinions in democratic societies. The
Internet has brought a whole range of new forms of resource effective participation.
These acts are usually on the smaller end of the spectrum, ranging from sharing and
liking images, texts and video clips to participating in online campaigns (Margetts
14
2019). Even though hashtags such as #blacklivesmatter, #metoo, and
#bringbackourgirls have gained enormous attention, protests and calls for collective
action are just a small part of this discussion. Often the purpose of the campaigns is to
gain attention, tell a story and express emotions (Highfield 2016). At the same time,
studies on political participation suggest that expressing political views online is a form
of slacktivism that rarely leads to other forms of political engagement offline. A less
cynical take is that political participation merely has moved into the Internet arena
(Vissers & Stolle 2013). Nowadays citizens sign petitions, communicate with officials,
and political groups in the virtual rather than physical world (Schlozman, Verba &
Brady 2010).
The Internet is also a key place for voters to interact with politicians and parties who in
turn are increasingly dependent on social media to get their messages out (Ernst et al.
2017). If social media is a platform for the uncensored opinions of the public (Ceron
2014), it is also a way for populist politicians to speak to the public without interference
from journalists (Ernst et al. 2017; Domonkos 2015): ”While the mass media adhere to
professional norms and news values, social media serve as direct linkage to the people
and allow the populists to circumvent the journalistic gatekeepers” (Engesser et al.
2017: 1110).
Many have raised concerns about the dangers to democracy in this new playing field, as
this dark side of social media reveals “fake news,” echo chambers, automated
propaganda accounts, voter manipulation, and dark ads directed at specific groups
(Benkler, Faris & Roberts 2018; Margetts 2019). Facebook collects user information in
order to target content to affect the desires, perceptions, and actions of the users. This
material can easily be taken advantage of by not only advertisers but by political actors
who wish to sway political opinion and impact voting behavior. Examples include
Cambridge Analytica and Russian election interference (Benkler, Faris & Roberts
2018).
On the most extreme end, we see groups such as Isis find sympathizers and recruit new
members via platforms such as Twitter, Facebook, and YouTube (Awan 2017). Right-
wing extremist groups use the Internet for mobilization and to strengthen internal
linkages (Caiani & Parenti 2013). At he same time, survey research in Sweden shows a
15
shift in opinion from support for the right to express extreme views online to a majority
supporting the limitation of that right (Internetstiftelsen 2018).
In tandem with this development, Twitter and Facebook have increased their restrictions
on which opinions users are allowed to express on their respective platforms (Dredge
2015; Gadde & Harvey 2018). Twitter has recently banned dehumanizing language on
their site, stating that: “Language that makes someone less than human can have
repercussions off the service, including normalizing serious violence” (Gadde &
Harvey 2018). Furthermore, people are being taken to court for posting
dehumanizing content on social media. A Swedish man was recently prosecuted for
hate speech after he compared Arabs with different animals (Iselidh 2018).
A study by Mellon and Prosser (2017) found that Facebook and Twitter users were
more educated and more liberal than the general population. The results were entirely
driven by a demographic difference in age, gender, and education. In consequence,
social media is in many countries not representative of the general population. This
year, 20% of Swedes have expressed political opinions online (Internetstiftelsen 2018).
This number overreaches the number of Swedish people who are members of political
parties by far. In 2011, this number had halved to 250 000 over the last 20 years
(Sandberg & Bjereld 2015). In accordance with previous research in other countries,
people who are younger, more liberal, as well as men participate to a considerably
higher extent than the rest of the population (Sandberg & Bjereld 2015;
Internetstiftelsen 2018). It is, nevertheless, the politically interested who express their
political opinions most frequently. The difference between politically interested a non-
interested in Sweden is 45% respectively 7% (Internetstiftelsen 2018). Although
bipartisan discussions exist, on a general level, users tend to interact with people with
the same political conviction (Jungerr 2016). The same is true in the Swedish case,
where people engage in discussions with people who share their political opinions to a
significantly larger extent (Internetstiftelsen 2018). These echo chambers of people
sharing the same political convictions, and reproducing the same content and ideas, are
feared as threats to democracy by some who believe that they lead to polarization and
foster extreme opinions (Margetts 2019; Benkler, Faris & Roberts 2018). The majority
of comments on all Party pages in Sweden is positive towards the party and show
support for their respective political proposals (Falasca, Dymek and Grandien 2017).
16
The reason can, of course, be that people tend to interact with those from the same
party, but there is also a possibility that the parties themselves remove negative
comments (Ibid). All parties in this current study demand a respectful tone in their
respective terms and conditions. The Moderate Party and the Social Democrat Party
also reserves the right to remove comments that include hateful or offensive language
(Moderaterna 2018; Socialdemokraterna 2018).
How politicians and political campaigns make use of Facebook and Twitter, and what
impact this has on voter behavior is the main discourse in the study of social media in
political science (Jungerr 2016; Spierings & Jacobs 2013; Zhang et al. 2010; McKinney
et al. 2014; Enli 2017). How citizens engage with politicians or express their attitudes is
understudied in comparison. Most studies research volumes of user data without
analyzing specific content. One common type of election result forecast is to count the
number of likes, followers and party interaction (DiGrazia 2013; Ceron et al. 2014;
Engesser et al. 2017). Studies rarely go beyond counting the number of likes when
evaluating citizen attitudes online. For example, Bobba et al. (2018) study a gender gap
in populist party support by counting the share of likes divided by gender. Evidently,
there is a need to go more in-depth in the analysis of citizen attitudes and behavior on
social media, as the study of how and which attitudes citizens express online is has been
left untouched.
In summary, previous research finds that extremism can grow and populist right-wing
voices are expressing themselves online, while the majority of Internet users are more
liberal and more educated than the general population. Platforms such as Twitter and
The opinion in Sweden has also shifted towards wanting to limit the ability to express
extreme opinions online. These conflicting results make predictions difficult for the
current study. Given that Twitter and Facebook prohibit dehumanizing language in their
respective terms of condition, and the probability that arguments unrelated to
dehumanization are the most central to the debate, we should expect few indications of
blatant dehumanization using animal metaphors. Given that this type of behavior can
result in anything from being banned from the site to being prosecuted, people probably
shy from using very stark language. Especially on sites such as Facebook, where there
is no anonymity.
17
2.3 Dehumanization – An overview This chapter is designed to describe dehumanization as a theoretical concept, and define
the intersections and differences with other phenomena in order to give clarity and
provide a foundation for building the analytical instrument. The first section gives an
overview of the field of dehumanization and the main strands of research. The rest if the
chapter outlines the concept and definitions of dehumanization, as well as the
relationship with other concepts.
2.3.1 Introduction to the Field of Dehumanization Given the multidisciplinary nature of dehumanization as a phenomenon, the concept has
previously been studied in multiple fields of research. These fields include psychology,
sociology, anthropology, communication, and to a limited extent within political
science. The predominant part of dehumanization research is, however, conducted
within social psychology (Haslam and Stratemeyer 2016). This strand will consequently
occupy the main part of this review. As this paper is a review of the use of
dehumanizing language, media depictions of the phenomena are also included in this
chapter.
Early dehumanization research focused on the context of conflict and mass violence
(Kelman; 1973; Opotow 1990; Bandura 1999). Dehumanization was studied primarily
as an enabler and a mechanism of genocide. It is a way to morally justify and distance
oneself from atrocious acts that would be unacceptable under normal circumstances
(Opotow 1990; Bandura 1999). There has been a significant proliferation of studies on
dehumanization within the last decade. The lion's share of these studies has been
conducted in social psychology and neighboring fields. Focus has shifted from the study
of dehumanization in intergroup conflict to everyday non-conflict settings where
perceptions of other groups are studied. Studies have shown that dehumanization is not
exclusively an inter-group phenomenon, as previously assumed, since individuals can
also be victims of dehumanization (Haslam 2006). Within social psychology, much
interest has been vested in studying the emotional response to different stimuli that are
said to trigger dehumanization and the effect that this has on attitudes and perceptions
of the dehumanized group (Haslam 2006; Haslam & Loughnan 2014). The majority of
18
these studies are experimental studies in laboratory environments where
dehumanization of fictive groups has been studied. A handful of studies have been
conducted where attitudes toward existing groups have been examined. These include
blatant dehumanization of Muslim refugees in Eastern and Southern Europe, as well as
disgust as a mediator of dehumanizing attitudes towards Roma people in Norway
(Bruneau, Kteily & Laustsen 2018; Dalsklev & Kunst 2015).
As previously noted, dehumanization is not just an outcome, but dehumanizing attitudes
can lead to support for violence, differential treatment and individual action against
other groups even in democratic societies (Maoz &McCauley 2008). For example,
people who dehumanize terrorists support the idea that the Swedish military should use
torture as a means to fight terrorism (Lindén et al. 2015). People who dehumanize are
also more likely to take action against immigration through protests and petitions and
support aggressive behavior against them (Bruneau, Kteily & Laustsen 2018).
Dehumanization also decreases willingness and propensity to help after natural disasters
(Andrighetto 2015; Cuddy 2007). Dehumanization can still affect the victim, even in
situations where it does not lead to other outcomes. Individual effects of
dehumanization have previously been studied through the experience of being
dehumanized and the effects on the dehumanized individual. Dehumanizing
maltreatment result in feelings of guilt and shame, as well as the feeling of lower social
status (Hartley & Fleay 2017; Bastian & Haslam 2011; Volpato & Contarello 1999).
Media and communication research, on the other hand, has focused on dehumanizing
depictions in news media. These depictions are predominantly in the form of animal-
metaphors or animal related words, disease and contamination, natural disaster or
pollution metaphors (Steuter & Wills 2010; Santa Ana 1999; Santa Ana 2002; O'Brien
2003a; Bleiker et al. 2013; Anderson 2017). In 2007, American paper the Columbus
Dispatch included a cartoon in which the map of Iran had been replaced by a sewer
swarming with cockroaches (Livingstone Smith 2011). In 2005 campaigns for British
Parliament, images of refugees as flows, floods or tidal waves were used as a way of
emphasizing not only sheer numbers but also how immigrants allegedly transform the
land that they sweep across in a damaging way (Charteris-Black 2006). In the United
States, Mexican immigration has been referred to as a "brown tide": a local metaphor
that connotes a brown algae infestation off the Gulf of Mexico (Santa Ana 2002).
19
During the 2016 presidential campaign, Donald Trump tweeted "Crooked Hillary
Clinton wants to flood our country with Syrian immigrants” (Utych 2018). These
statements should not be discarded as mere rhetoric. There has been some cross-
fertilization between social psychology, and media studies, in the study of
dehumanization. Previous studies have shown that negative portrayals of immigrants
impact dehumanizing attitudes within the audience (Esses, Medianu & Lawson 2013).
In turn, dehumanizing depictions of immigrants lead to negative attitudes towards
immigration (Utych 2018; Goméz-Martínez & de la Villa Moral- Jiménez 2018).
2.3.2 The Concept of Dehumanization Many people are surprised to hear that dehumanization has not been widely studied in
social science, chiefly because the word is often used as a tool in debates, especially
surrounding sensitive issues (Livingstone Smith 2011). Dehumanization was mentioned
10- 30 times per year by Swedish newspapers between the years 1960 and 2012, while
in 2015 the usage peaked at 794 times (Svenska Dagstidningar 2018). At the same time,
clear definitions, systematic analyses, and supporting evidence are lacking both in the
debate itself and the study of it. This conclusion is further supported by Haslam et al.
(2008a) who emphasize how the myriad of studies under the umbrella of
dehumanization fail to show when dehumanization can be claimed. This suggests that a
systematic and transparent approach to dehumanization is needed, especially in social
science.
Dehumanization can be expressed in different ways: discursively, in the treatment of
others as well as experienced. Regardless of which, it stems from the perception of the
other as subhuman (Livingstone Smith 2011). Most scholars agree that dehumanization
should not be seen as a dichotomous concept where dehumanization is either absent or
present. It usually described as working on a continuum or on a discrete scale (Oliver
2011; Livingstone Smith 2011; Savage 2013). This assumption is adopted across the
board, in social sciences, psychology, and neuroscience.
Dehumanization has been described as a spectrum from mild to severe, or subtle to
blatant. When it comes to dehumanizing actions, genocide is usually recognized as the
most severe expression of dehumanization (Livingstone Smith 2011), while social
20
exclusion and slight degradation are considered as milder forms (Haslam & Loughnan
2014). However, an important question, which every scholar much address, is from
where this scale departs, and whether infrahumanization – the denial of uniquely human
emotions (Leyens et al. 2000) is recognized as a form of dehumanization. The problem
can partly be attributed to a lack of definitions, which we will return to later in this
chapter. Another reason is the lack of clarity regarding what is meant by the terms
subtle and blatant dehumanization. Usually, these words refer to either a perception of
the other or how the other is described, rather than dehumanization actions.
There is no consensus among dehumanization scholars on what is meant by subtle or
blatant dehumanization, and the terms are often used without being defined. Whether
this is a question of degrees of overtness, or whether someone is partly or fully
dehumanized is not always clear. In order for scholars to build descriptive theories on
dehumanization, it is important to clarify what is meant by these terms. The distinction
is also important to make since the produced effect is different depending on the form of
dehumanization. Whereas distinctive feature of subtle forms of dehumanization is that
they can permeate social relations regardless of intergroup conflict, or the knowledge of
the actors involved, blatant dehumanization is direct, explicit and overt. The latter is
said to be more closely connected to aggressive attitudes, support for torture and violent
action towards other groups (Volpato & Andrighetto 2015; Kteily et al. 2015). By using
distinctions made by Haslam and Loughnan (2014), and Kteily and Bruneau (2017)
respectively I have created a matrix of different examples of discursive dehumanization
to clarify the concept, and to illustrate the differences between different types of
dehumanization.
21
Figure1:TheSpectrumofdehumanization
Thisfigureshowsdehumanizationonascalebetweenimplicitandexplicit,respectivelypartialandfull
dehumanization.Themathematicaldistancebetweentheexamplesisnotanaccuratedepiction;itis
ratheranillustrationoftherelationship.
Many scholars use the term subtle dehumanization to refer to a partial exclusion from
humanity; others use it to signify that the dehumanization is implicitly conveyed
(Haslam & Loughnan 2014). When a group is partially excluded, they are deprived of
certain aspects of their humanity such as agency, and when a group is fully excluded,
they are deprived of their full humanity (Figure 1). When Bruneau et al. (2018) use the
term blatant dehumanization they refer to expressions that are both overt and absolutely
dehumanizing, such as calling someone a subhuman, pests or vermin (Figure 1, upper
right corner).
Not all animal metaphors are fully dehumanizing. Calling someone an Ape is a way of
explicitly denying a person some human characteristics such as cognitive abilities
(Figure 1, upper left corner). The reason is, that humans and several other apes share a
closer taxonomical relationship than lets say humans and rats. Animal metaphors with a
more distant relationship to humans are consequently an explicit and more absolute
22
exclusion from humanity. A group that is depicted as pests or vermin is being denied
both cognitive and emotional abilities, while a person depicted as an ape is only being
denied cognitive abilities (Tipler & Ruscher 2014).
Using animal-related words to describe the actions and behavior of humans is an
indirect, yet still a way of fully dehumanizing someone (Figure 1, lower right corner).
Examples of such sentences are ”Suspected al-Qaida Nest”, ”Muslim Sect Breeding
ground for Al- Qa’ida” and “public benefits are a lure to immigrants” (Steuter & Wills
2010: 155-156; Santa Ana 1999). By using terminology that is associated with animals
rather than humans of the 21th century their inhumanness is implied.
This study will predominantly study forms of dehumanization, which are partially
excluding, and on the implicit end of the dehumanization spectra, as it is highly unlikely
that blatant forms of dehumanization exist in moderated areas of social media. This
form is often referred to as subtle dehumanization (Figure 1, lower left corner). This
occurs when a group is implicitly deprived of one aspect of their humanity, by for
example being called immoral, irrational or unrefined (Haslam 2006).
2.3.3 Defining Dehumanization
Conceptual discussions are often lacking in the study of dehumanization especially in
social science. In many cases, definitions are often implied or simply assumed without
argument. One example is an otherwise ambitious study by Haagensen & Croes (2012)
where the empirical indicators used in their study is a compilation of dehumanizing
manifestations found in other studies. Steuter and Wills (2010) use the same approach
in their analysis of media discourse and the war on terror. They depart from previously
established animal metaphors without defining dehumanization.
In sociology, dehumanization has been defined as the essentialization and denial of the
humanness of a group, which is subsequently excluded: ”a denial that a certain group is
‘equally’ human, no matter how ‘humanity’ is defined” (Savage 2013:143). In this case,
the door is left open for all possible meanings one might put into the word
23
dehumanization. By adding the term ‘equally’ human, all forms of partial and verbally
subtle dehumanization are included. This definition is consequently rather wide.
By not defining who belongs to the realm of humanity, or what makes a human we do,
however, open up for arbitrary categorizations of dehumanization. In addition, if we are
interested in studying aspects of dehumanization beyond self-explaining dehumanizing
metaphors, we also need to define what humanness is. Otherwise, we cannot study
dehumanization in negative terms, i.e., what someone is not. These definitions are
lacking across the board with only a few exceptions (Haslam 2006; Savage 2013).
Others apply stricter definitions of dehumanization, which do not encompass partial
dehumanization: "The act of conceiving of people as subhuman creatures rather than as
human beings" (Livingstone Smith 2011: 37). In this case, a person must be perceived
as something that is truly subhuman. It is not sufficient to be for example treated as an
object (Livingstone Smith 2011). These narrow definitions certainly serve the purpose
of not watering down the concept. However, the problem is that this definition only
allows us to study dehumanization in contexts marked by extreme violence and
atrocities. The everyday and less blatant expressions of dehumanization are lost if we
use this definition as a measurement.
In order to study the less direct forms of dehumanization, we need a definition that
encompasses subtle forms of dehumanization. Therefore this study adopts a definition
widely used in social psychology. This study defines dehumanization as the act of
perceiving someone as less than fully human (Haslam and Stratemeyer 2016: 25).
Seeing someone as less than human can serve as a way of legitimizing dehumanizing
practices (Haslam & Stratemeyer 2016). Within contemporary social psychology “less
than” human usually implies a scale of humanness rather than a dichotomous
relationship. Someone can be subhuman, non-human or less than human compared to
others. If we accept that “several features are consensually considered necessary
(though none of them is sufficient) to be perceived as human beings” (Leyens et al.
2000: 187), we can agree that being denied one of those qualities is dehumanizing. In
other words, the quality of being a rational person is not sufficient to qualify as a
human. In that case, artificial intelligence could be human. However, rationality is one
of the features, which are considered necessary to be called a human.
24
2.3.4 Relationship to other Concepts
Is dehumanization rendered superfluous when we have concepts such as prejudice,
racism and ethnic discrimination? The answer is no. In this section I will explain why
by distinguishing dehumanization from other related concepts.
Since dehumanization is often used as a tool to legitimize differential treatment and
segregation, it can be used for purposes of ethnic discrimination and subjugation of
women (Haslam 2006; Volpato & Andrighetto 2015). Racism as an ideology is built on
the idea that different groups of people can be divided into subgroups based on their
alleged qualities, characteristics, and capabilities. Patriarchal ideas are based on the
notion that women are inferior to men. Dehumanization is also an idea based on the
belief that there are vertical divides between groups of people. Hence, the main
similarity is that they are methods to debase other groups. They are, however, different
constructs. For example, dehumanizing metaphors attain a sexual dimension when
women are depicted as "prey" such as bunnies, which distinguishes it from the common
prey metaphor where men are portrayed as game animals. By attributing such
metaphors to women it implies that these women are flirty temptresses that are easily
dominated and passive (Tipler & Ruscher 2017). The implication of the female prey
metaphor is that women should be domesticated. This metaphor can consequently be
seen as a justification of the subjugation of women. However, sexism can occur without
dehumanization and vice versa. Women can be treated as objects and simultaneously
being seen as human (Livingstone Smith 2011). The same goes for dehumanization
based on ethnicity, sexual orientation and of people with disabilities.
Although prejudice and dehumanization appear simultaneously in some cases, they too
are conceptually different (Bruneau, Kteily & Laustsen 2018; Kteily et al 2015; Harris
& Fiske 2011). Prejudice is defined as negative attitudes towards a person due to them
belong to a particular group, for example, a sexual minority (Hodson & MacInnis
2012). The common denominator between dehumanization and prejudice is that they
both involve negative sentiments and involve intergroup bias. However, prejudice can
involve ascribing someone with any trait, while dehumanization specifically involves
the denial of human traits. The central tenet of prejudice is the negative sentiment, while
the main component of dehumanization is the rejection of humanness (Hodson &
25
MacInnis 2012). It is possible to strongly dislike a group of people on account of a
stereotype without dehumanizing them. Equally, a person can be dehumanized in the
absence of negative prejudice. One such example is the image of the “Noble Savage”
who symbolizes goodness while lacking civility and refinement (Bruneau, Kteily &
Laustsen 2018). Accordingly, the sentence "vulnerable people lacking education and
knowledge of the Swedish language have no other option but to beg" is not
dehumanizing. The reason being that it is permeated with empathy and understanding,
rather than negative sentiments.
To conclude this chapter, state of the art dehumanization research can provide some
indications on the direction of the results of this study. Dehumanization is conceptually
distinct from other related phenomena. It should be regarded as a spectrum where
dehumanization can take different forms, and be expressed in a multitude of ways. This
study examines discursive dehumanization in particular, and examines both explicit and
explicit forms, with a focus on the latter. Given that subtle dehumanization, where
someone is implicitly and/or partly denied their humanness is more widespread in
everyday contexts, we are more likely to find evidence of this rather than explicit forms.
Previous studies show that dehumanizing views exist on a societal level and can be
transformed into action. Still, there has been no study of the prevalence of subtle
dehumanization in political debate. As for this study, previous research suggests that we
should be able to find dehumanizing language in political discussions, but it does not
give us an idea of the extent to which it is being used.
26
3 Methodology This is a quantitative study designed to map the extent to which dehumanization is used
in political discussions online. This chapter presents and discusses the methodological
choices and problems of this paper starting with design, method, and material; ending
with operationalization, including the analytical tool.
3. 1 Design
Even though subtle dehumanization has been extensively studied in other academic
fields than political science, this particular method and tool for analysis have not yet
been tried. Before the research was initiated there were few indications to whether we
would find any evidence of discursive dehumanization. Likewise, it is difficult to
generate testable hypotheses from a single case study; therefore a simple case study
design was chosen for the study at hand. Even though the definition of a case study has
been described as a "definitional morass" (Gerring 2013: 1136), it is despite this a
useful way to capture how this study investigates one phenomenon with the help of one
political issue.
3. 2 Method
The chosen method is a quantitative and descriptive content analysis of social media
discussions on street begging. The purpose of the analysis is to describe the substance
characteristics of social media content. Finding indications of more subtle forms of
dehumanization is a challenge, but in order to bring stringency and predictability to an
otherwise eclectic field in social science, it is preferable to turn to quantitative methods.
For the same reason, I'm using predefined categories in the analysis. Using quantitative
methods also allow us to build on extensive dehumanization research in Social
Psychology, and text analysis, in particular, complements previous research in that field.
This study seeks to answer whether and how dehumanization is used in political
discussion on social media. Therefore, the frequency and type of dehumanizing
messages is the focus of the analysis. Since the content analysis is theoretically driven,
27
the categories are derived from previous literature on dehumanization. You can find
more information about the variables and categories in the chapter on
Operationalization. Traditionally, units of analysis are coded with the help of a physical
code sheet (Esiasson 2017). In this instance, I coded the material through Microsoft
Excel. You will find the Codebook in Appendix 2.
Quantitative content analysis is indeed designed to be systematic and objective
(Esiasson 2017). The extensive analytical tool (Appendix 1) and the exemplification of
the coding process (Appendix 3) are some of the ways in which I tried to achieve this.
Still, the superior way of ensuring high reliability is by having multiple coders and to
conduct intercoder reliability tests. Due to lack of resources, it was not possible to look
at the agreement between the results of independent coders. The fact that I am the sole
coder should be taken into account when assessing the reliability of the results of this
paper.
3.2.1 Case Selection Being a humanitarian nation is a central part of Swedish foreign policy, and the country
has been called the last bastion of liberal values in Europe (Embury-Dennis 2018;
Stavrou 2018; Al Jazeera News 2018). Thus, if we find evidence of dehumanization in
Swedish everyday politics, we are likely to find it elsewhere as well. This evidence
would also provide strong evidence for the existence of dehumanization towards
immigrants in particular and dehumanization in everyday politics general.
Applying the same logic in case selection is ill-advised as this is the first study of its
kind and we need to use a case that lets us try the framework of analysis.
Dehumanization is expressed through perceptions or treatment of other groups and is
predominantly an intergroup phenomenon. The issue of street begging is consequently
suitable, as it is centered on the perception of a group of people that exist inside, but
apart from Swedish society. Since the issue and the group are closely connected the case
provides a good foundation for trying out whether it is possible to conduct a study on
discursive dehumanization. This is especially important when it comes to material
selection because it allows us to use a neutral term such as begging (tiggeri in Swedish)
and still capture a central proportion of the debate. Consequently, this is a matter of
28
weighing feasibility and the possibility of being able to conduct this study, against
external validity. Given that begging is such a contentious issue in Sweden,
generalizations to other political issues should be made with caution.
3. 3 Material Getting first-hand access to the space of everyday politics is not easy; as it takes place
outside the traditional political sphere and in the informal everyday of citizens
(Highfield 2016). It would require us to observe political debates at breakfast tables, in
the office, and private associations. Luckily, social media provides a window into
everyday politics, as it has become the center stage for people to express their private
opinions publically. 72% of all Swedes use Facebook, 60% use Instagram, 38% use
Snapchat, and 22% use Twitter (Davidsson, Palm & Mandre 2018). Even though
Instagram provide the opportunity to comment and engage in political discussions, it is
rarely used as such. For example, the search term "beggar" (tiggare in Swedish) almost
exclusively generates pictures of cute dogs begging for food. Which makes Facebook
and Twitter the most suitable platforms for analyzing political opinions.
Even though a large proportion of the Swedish population use Social Media, we cannot
claim that this group is representative of the general population, as Internet users and
non-internet users differ in some important regards. In the case of Sweden as well as
other countries, Internet users are less authoritarian and more positive towards
immigration than the general population (Sandberg & Bjereld 2015; Mellon & Prosser
2017). It is also important to keep in mind that the discussant in this study might differ
from the general population because they chose to partake in political discussions,
something that not all citizens do.
The selection of sources was based on several criteria. Accessibility and representation
were important factors. In order to avoid bias in the selection, the material had to
encompass both sides of the debate and not include forums where we are most likely to
find dehumanization. Consequently, certain popular forums on the fringe such as
flashback and Swedish far-right forum Samtiden (formerly known as Avpixlat) were
not an option. In addition, forums that offer anonymity tend to have a harder tone than
for example Facebook (Halpern & Gibbs 2013). Large social media platforms such as
29
Twitter and Facebook have terms of use that does not allow hateful language. Twitter
has now moved even further to ban dehumanizing language in particular (Gadde &
Harvey 2018). This is not necessarily a problem for this study, as it focuses on mapping
subtle dehumanization.
Virtually every Swedish newspaper has removed the ability to comment their articles
online, as well as social media. The reason being the difficulty to moderate the online
forum to ensure that the discussion does not escalate, especially in discussions around
sensitive issues such as immigration (Granström 2018; Aftonbladet 2014). Which
restricts the possibility to use comments on news articles as sources.
The chosen unit of analysis is comments and posts from Facebook and Twitter. As these
comments are predominantly one sentence long, we do not have to worry too much
about pulling sentences out of context. Furthermore, some contextual interpretations
had to be made regarding whether the comment is an answer to another comment and
the current discussion topic. For example, if a person writes, “no [name of another
discussant], you are wrong,” it is interpreted as a positioning against the discussant
whom they referred to in their comment.
3.3.1 Data collection For purposes of external validity, the optimum method would be random sampling.
However, the limitations of the respective social media platforms do not allow this,
which propels us to use other methods of selection. Facebook has become extremely
restrictive on handing out large amounts of data after the Cambridge Analytica scandal,
which makes automatized data collection very difficult. Hence, the information will be
manually collected from party Facebook pages. When this study commenced, and
before the data was analyzed there was no way of knowing whether we would find a
sufficient amount of dehumanizing comments to study development over time, which is
why the study is limited to one period. The time period from which data was gathered is
2017-2018, which is the time frame of the current debate. Consequently, the parties that
are represented in this study are the ones that have been vocal about the issue on social
media. This selection coincides with those who have played the most active part in the
debate: Sweden Democrats (Sverigedemokraterna, SD), Social Democrats
30
(Socialdemokraterna, S), and the Moderate Party (Nya Moderaterna, M). The data set
created for this study contains a total of 586 comments, where 461 of these are
Facebook comments (167= Sweden Democrat, 137= Social democrat and 157=
Moderate Party) and 125 are Twitter comments. In contrast to private Facebooks
groups, a public party page is a discussion between parties and citizens thus a much
more open and accessible forum for discussion. The selection of posts was based on
which were most discussed (amount of comments, likes, and reactions). Therefore we
are less likely to find dehumanization here, which means that the conclusions derived
from this paper can represent a baseline in private forums.
I used the same period for Twitter as Facebook and made sure to disable the quality
filters before collecting the data. Due to Twitter search metrics and filters, it is
complicated, if not impossible to collect all the tweets that exist on a specific topic.
Consequently, it is hard to know the sheer size of the population of tweets from which
we draw our sample (Lorentzen & Nolin 2017). Using the same method of selection for
Twitter as I did for Facebook is not feasible, as people interact with specific politicians
rather than parties, which would make the selection method too time-consuming. A
term such as begging is not exclusively used in the debate around this particular issue,
but other discussions as well. Therefore, I chose to use the most popular hashtag to
discuss Swedish politics #svpol and combined it with the term "beggars" OR "begging".
All comments in the search were collected for analysis. By using this hashtag, we only
find comments from people who are interested in having a political discussion.
Moreover, politicians, journalists, opinion makers, and news outlets are using the
hashtag as well. This suggests that it is central to the debate, and not on the fringe or an
extreme end of the political spectra.
Afterwards, the collected data was filtered according to criteria of relevance: Whether
the statement or comment is addressing the issue of begging or not. For example: "if
you collaborate with party x you will win the election", is excluded accordingly.
Comments that are related to other issues will not be included. Examples of non-
relevant comments are those who slander other political parties, personal attacks on
other discussants, tagged friends etc. Spam was also filtered away in the coding process.
Comments posted several times were counted as one in order to avoid a skewed result
(Ceron et al. 2014).
31
3.3.2 Ethical Considerations Research on Social Media is a developing social science field, and naturally articles,
which deal with ethical risk and dilemmas, are few (Moreno et al. 2013). A pivotal
question that research on Social Media needs to answer is whether it involves human
subjects or not. In order to not qualify as research without human subjects, the
following criteria need to be met: “access to the SMW[Social Media Website] is public;
information is identifiable, but not private; and information gathering requires no
interaction with the person who posted it online” (Moreno et al. 2013: 16). Regarding
the current study, the information is publically available on the Party pages and Twitter
hashtags. I did not interact with any of the subjects, nor the Facebook pages, or collect
any of their personal information. The units of analysis in this study are the comments,
not the users. As I have not obtained consent from the discussants, I chose to exclude all
personal information and personal identifiers when giving reference, as gender and
Party page. This is a small trade-off given that this information is interesting, but not
imperative for the reader to acquire.
3.4 Data analysis
This section builds on the overview and conceptualization of dehumanization from the
theory chapter and contains the method of analysis: operationalization, indicators and
analytical framework.
3.4.1 Operationalization
Psychologist Nick Haslam’s theory of dehumanization was used as a basis for the
operationalization. His theory is the most highly developed theory as it has sprung from
empirical research and has been tested empirically by different scholars. It is based on
an extensive literature review in which he analyses different conceptualizations, and
subsequently presents two forms or facets of dehumanization (Haslam 2006). This
disaggregation has been corroborated by subsequent research on dehumanizing
perceptions and experiences (Bastian & Haslam 2011; Andrighetto et al. 2014; Hartley
& Fleay 2017). The two forms are derived from the idea that there are two distinct ways
to define humanness. One definition states that there are distinct characteristics that
32
make us human, i.e. typically human traits such as agency, individuality, emotional
responsiveness, cognitive openness and interpersonal warmth. The other definition
postulates traits that differentiate humans from animals, i.e. uniquely human traits such
as civility, refinement, moral sensibility, rationality, and maturity (Haslam 2006). These
aspects of dehumanization are featured in the table below, including ways in which
these attributed can be denied. By using this theory we are also able to study subtle
expressions of dehumanization, where someone is partly, or implicitly dehumanized.
Table from Haslam (2006:257)
Defining humanness according to human essence and what makes us human easily
leads to subjective judgments and normative assumptions about what the core of a
human being is. In addition, dehumanization focused on the denial of human essence is
more closely associated with the dehumanization of groups other than immigrants. For
example, depersonalization and emotional non-responsiveness can characterize doctor-
patient relationships, and objectification of women generally involves a denial of
agency and individuality (Haque & Waytz 2012; Tipler & Ruscher 2017). Whereas the
denial of characteristics that differentiate humans from animals is related to vertical
hierarchies and lower status (Bastian & Haslam 2011; Haslam 2006). Negative
stereotypes of Roma people depict them as lazy, unhygienic, criminals, with hostile and
aggressive behavior (Orosz et al. 2018), which are all associated with animalistic
dehumanization. Therefore, I have chosen to restrict this inquiry to examine the
dehumanization of uniquely human traits. The research by Orosz et al. (2018) on Roma
in particular, as well as research on the dehumanization of immigrants in general (Esses
33
et al. 2008), leads to the careful expectation that moral dehumanization will appear in
the results.
3.2.2 Operational Indicators
Drawing from Haslam’s definition of human uniqueness I have created different
indicators for subtle dehumanization. Since these aspects of dehumanization are
common adjectives used to describe someone, dictionaries along with previous research
on dehumanization have provided definitions and examples. The analytical matrix
(Appendix 1) was construed prior to the analysis, and only downgrading examples were
used. The variables coarseness and lack of culture are difficult to separate both in theory
and practice. Some dictionary definitions overlap, where lack of manners or
impoliteness is sometimes used to define both adjectives. Politeness and cleanliness
were thus connected to lack of culture, while aggressive and offensive behavior was
classified as coarseness before the initiation of the study. However, as the analysis
progressed, it became evident that coarseness and lack of culture are difficult to separate
also in practice. In the analysis, the two variables were therefore aggregated. Even
though previous research in dehumanization displays several examples of so-called
moral dehumanization, where immigrants are described as bogus, and as cheaters, I
have complemented this by using a breakdown of the rules of common morality. These
components are "do not deceive," "keep your promises," "do not cheat," "obey the law"
and "do your duty" (Gert 2004). For full operationalization, and examples see
Appendix 1. Appendix 3 provides examples of how the analysis was conducted in order
to be transparent and give the reader insight into the analytical process.
To capture even more explicit aspects of dehumanization, the categories of animal
metaphors and animal behavior was included in the analysis. Since indications of these
metaphors were few, the variables were aggregated into one the coding scheme. They
are perhaps not as prevalent, but they represent a stark and deliberate form of
dehumanization. Although some animals represent the same characteristics that are
subtly denied someone, such as cognitive abilities, they are still different categories, as
they are linguistically separate expressions. In addition, the less direct type where
someone is denied human characteristics can be expressed without the awareness of the
perpetrator (Volpato & Andrighetto 2015).
34
3.2.3 What is not Dehumanization?
In the spirit of good research, it is important to clarify the aspects, which are not
indications of dehumanization. Using animal words does not automatically qualify as
dehumanization. A person is not automatically dehumanized simply because an animal
metaphor is used (Haslam et al. 2008a). A vast amount of idioms and proverbs in many
different languages use animal metaphors to highlight positive qualities in people.
Expressions such as: "strong like an ox", "as busy as a bee", "night owl" and "top dog"
are consequently not dehumanizing. This is because the meaning of the animal word is
replaced by another meaning in an idiom.
An example of moral dehumanization is “you see beggars enter a fancy Mercedes at the
end of the day, so they are definitely not poor, they’re just want to trick you into
thinking that they are", while a sentence stating that "sometimes, the money that they
earn go to trafficking" is not dehumanizing. The difference is whether beggars are
considered immoral, or whether they are being used. In a similar manner, saying that
begging is organized is not dehumanizing in the same way as saying that begging equals
organized criminal activity. As many negatively charged words have positively charged
counterparts or similes, the former is usually chosen in the context of dehumanization.
In the matrix, I only used negatively charged sentences. During the process of analysis I
double-checked I consulted dictionaries to confirm whether some words are considered
downgrading or not.
Statements that are negative towards a group are not necessarily dehumanizing. "They
don't belong here", or "stop imported begging", cannot be classified as dehumanizing
unless there is a reference to why they don't belong here, or why imported begging is
different from so-called domestic begging. The underlying reason could be
dehumanization, but not necessarily. Other examples from this study are: “Ban begging
completely, [I’m] so tired of seeing them at every grocery store, by the subway etc”
(Appendix 3). Being tired of seeing suffering might be an effect of dehumanization, but
we cannot determine that from the statement itself. Even aggressive statements such as
"I am so angry at begging, I just want to shove a pizza in their faces" or "we should all
stand up and go to every damn store where there's beggar scum" are not dehumanizing
on their own.
35
4 Results and Discussion The question that this paper sets out to answer is whether and how dehumanization is
used in the debate on street begging in everyday politics. The issue was mapped using a
novel tool for content analysis, which contained five elements of dehumanization that
were based on previous research in social psychology. These elements are lack of
culture, coarseness, amorality, irrationality and immaturity. In this section, the empirical
findings from the quantitative content analysis are first presented, and thereafter
discussed in the subsequent section.
4. 1 Results The results section starts with an outline of the general results, where the overall share
of dehumanizing comments is presented. After that, the chapter continues with a
breakdown of results by party page, and lastly, dehumanization is disaggregated into
different elements in order to show how dehumanization is used. Issues and insecurities
regarding the tool of analysis are also explained in this section. Whereas the analysis
was conducted in Swedish, the quotes included in the results section are in English and
translated by me. As explained in the ethics section, I chose not to include the names or
aliases of the people who engaged in the discussions, even though they are publically
available.
4.1.1 General Results After removing spam and comments that did not concern the debate, the total amount of
dehumanizing comments in all categories were 130 out of 586 (Table 1). Which means
that the aggregated amount of dehumanizing comments add up to 22% of all arguments
used. All discussants that used dehumanizing arguments also expressed negative or
hostile attitudes towards beggars and begging. Even though most commentators
expressed their dislike with beggars, only a handful explicitly stated that they want a
ban on begging: "Begging should be banned! We're not going back in time. So
annoying with all these beggars and they have no self-control, especially not when they
sit outside the regional hospital where people who are seriously ill have to put up with
36
these people every time they seek care". 1 Others used dehumanizing arguments not to
support a ban, but to support the removal or deportation of people who beg: “[…] If
someone likes begging, trafficking and prostitution they should be deported along with
these parasites […]”.2
There was no noteworthy difference between the share of dehumanizing comments on
Facebook and Twitter respectively. However, within the material, there are differences
in the percentage of dehumanizing comments depending on which party page the
comment was posted on. Unlike Facebook, the Twitter results were not disaggregated
by party, since only a few cases were clear-cut. Among the users in this material,
ideological preferences were more often stated. Some users also shared material and
used hashtags from several parties, which created coding issues. Other differences that
this study tested for was a difference in gender. When doing a simple correlation there
was no relationship between the variables gender and dehumanization in this sample.
Table1Frequencyofdehumanizingcomments:Numberofcommentscontainingatleastoneformof
dehumanization,andnon-dehumanizingcommentsrespectively
4.1.2 Party page Specific Results Figure 2 shows the rate of dehumanizing comments for each party page. Even though
we cannot know for sure whether discussants on party Facebook pages are supporters of
those parties, we can observe a discrepancy in the share of dehumanizing comments
between different party pages.
1Nr.3242Nr.67
Dehumanizing Non-dehumanizing Total
Facebook: SD 72 95 167 Facebook: S 7 130 137 Facebook: M 18 139 157 Twitter 33 92 125 TOTAL 130 456 586
37
Figure2Dehumanizationbysource:Shareofdehumanizingcommentsforeachsource.SD=SDon
Facebook,S=SonFacebook,andM=MonFacebook.
There was no significant difference in the percentage of dehumanizing comments
between the respective social media platforms. The share of dehumanizing comments
on Twitter was 26%, while the total Facebook average was 22%. The lowest share of
dehumanizing comments, 5%, was found on the Social Democrat Facebook page, where
only a few individual comments contained dehumanization. It is important to point out
that a majority of the discussants where against a ban on begging, stating that such a ban
would de facto be a prohibition of poverty. This needs to be underscored because almost
all dehumanizing comments were used to depict begging and beggars in a negative
light. The second lowest share was found on the Moderate Party Facebook page on
11%. The Sweden Democrat page had the highest share of dehumanizing comments
with 43%.
Without further studies we cannot draw any conclusions based on this discrepancy, we
can only note that it is there. However, many people openly expressed their support or
political identity through comments in the style of “That’s my party <3” or “that’s not
the Social democrats I voted for.” Which indicates that at there is at least some
correlation between commenting on a party page and actual party support.
0% 5%
10% 15% 20% 25% 30% 35% 40% 45% 50%
Facebook: SD Facebook: S Facebook: M Twitter
Dehumanization by source
38
4.1. 3 Elements of Dehumanization
The elements of dehumanization that are mapped in this paper are lack of culture,
coarseness, immorality, irrationality, and immaturity. The majority of the dehumanizing
comments were partly dehumanizing, where specific qualities were denied people,
rather than their complete humanness. Only 6% of the dehumanizing comments were
based on explicit metaphors or metaphors that entirely exclude people from the realm of
humanity; the words “Parrots3” and “parasites4" were the only examples of animal
metaphors in the material. Other examples of more explicit dehumanization were
garbage related metaphors such as: "A ban is a good thing. We don't need to be the
landfill of Europe".5 It is possible that other explicit dehumanizing comments have been
removed by the moderators of the party pages, or have been reported by other users.
Unfortunately, we cannot know whether that is the case, or to what extent comments are
removed, or who removes them.
The analysis conveys little variety in the elements of dehumanization that were used. As
shown in Figure 3, the results show a dominance of mainly two elements of
dehumanization: immorality with 56% and the combined category uncultured/unrefined
with 34%.
Figure3Elementsofdehumanization:Percentagesofdifferentelementsofdehumanizationofthetotal
dehumanizingcomments
3Nr.46andNr.474Nr.39andNr.675Nr.278
34%
56%
3% 1%
6%
Elements of Dehumanization
Uncultured/unrefined
Immoral
Irrational
Immature
Animals etc.
39
Even though there is a dominance of mainly two elements of dehumanization, there is
considerable variety within these two categories. What they have in common is negative
sentiments towards beggars. The next section contains a presentation of the results from
the main categories, including further classifications of the different components of
dehumanization based on the analytical tool.
Questioning Civility and Refinement Questioning the civility or refinement of beggars was the second most common strategy
of dehumanization. Claims that beggars are rude, engage in harassment, or are
unhygienic are the most frequent comments of this type.
As previously explained in the methods section, I chose to aggregate the categories of
lack of culture and coarseness, as they are overlapping in theory and in practice. The
problem with using a combined category is that it contains several different elements,
which can be impractical. Especially when a concept such as civility can refer to things
on different dimensions such as physical appearance as well as behavior. I chose to
aggregate these categories despite these issues, as they fill no functions separately
without clear distinctions. The next examples will show that line between impolite and
crude behavior is not always easy to draw. In many cases, being rude and offensive are
parts of the same narrative where beggars negatively affect the social environment:
“[…] I think it is rather deplorable that I’m not able to sit on the subway or
commuter train without being molested by obtrusive beggars. They don’t pay for
their tickets yet they are allowed to harass us who actually have”6
A person who is obtrusive is “forward in a manner of conduct” with no regard for
propriety (Merriam- Webster). This type of behavior can be interpreted as both impolite
and untactful, especially in a Swedish context. The comment would thus qualify as
depriving someone of civility. It can also be described as aggressive behavior. In that
case, the comment would qualify as crude.
6Nr.26
40
Another way of depriving someone of civility and refinement is to picture them as filthy
and unhygienic. These claims are often used to support a claim that beggars negatively
affect the physical environment in terms of cleanliness and sanitation:
“Roma beggars threaten park workers who have to clean up the beggar poo after
them”7
In some cases, anecdotal evidence paints the picture of beggars as negatively affecting
both the physical and social environment:
“It is hardly the Salvation Army or the Church who create problems. I have never
met e.g. a Salvation Army soldier who runs around in the subway shoving rattling
cups or newspapers in my face. Neither have I heard about priests dirtying parks or
campsites.”8
The element of degradation, which is a fundamental part of dehumanization, comes out
very clearly when a group is contrasted against another. This comment is not the only
example of beggars being dehumanized by being negatively compared to other groups
such as charity organizations, people with traditional jobs, or Swedish people in general.
Questioning Morality
Moral dehumanization of beggars represented a majority of dehumanizing comments
with 56% of the amount. The different varieties and ways of questioning the moral
capacity of beggars largely corresponded to Gert’s (2004) definition of the rules of
common morality. A person violates these rules by not obeying the law, do their duty or
keep their promises, and by being deceptive and cheating (Gert 2004). The only aspect,
which was not included among these, was keeping one’s promises.
The were many examples of discussants equating beggars with criminals, or claiming
that beggars disrespect Swedish legislation. Some examples include calling beggars
“Bandits,” “Criminal trash,”9 or claiming that “they create problems by committing
unquestionably criminal acts.”10 A common comparison found in the commentary was
between beggars and charity organizations, where the latter is described as law-abiding
7Nr.5588Nr.3939Nr.16710Nr.383
41
and always seeking a permit for collecting donations in public, while beggars are
contrasted as violators of the law who never bother to acquire permission to sit
anywhere.
A typical narrative is that beggars “cheat the system” by not paying taxes like
everybody else, and they that are not actually in need because they are not actually poor,
or because they earn a lot of money from begging:
“I saw a beggar woman I recognized get into a fancy Mercedes… Could have
been the mob, or maybe her husband… They were not poor that’s for sure, not
like our pensioners.”11
Another common way to undermine the needs and vulnerability of beggars is to claim
that a poor person would not have a cell phone, ergo beggars cannot be truly poor.
These arguments are sometimes coupled with the claim that beggars are deceptive, and
will try to deliberately trick you, or take advantage of your good will:
“A thing with beggars is that they are hardly poor, some of them have insanely
expensive cell phones, and they pretend to have one leg and then they often have
two when they get up and nobody is around. So stop feeling sorry for them. They
are just making fools of us all”12.
The material contained several other examples such as the one above, where it is
claimed that beggars fake disabilities to deceive you and take your money.
The related idea that beggars “do not do their duty” was illustrated by portraying them
as idle and lazy. The notion that begging is not “honest work”, is prevalent in the
material: “I told a beggar the other day “do you really think that I should give you
money that I’ve worked hard for, when all you do is sit on your ass all day long and
don’t do anything to support yourself”? She had no answer to this.”13 This comment
reflects the attitude that begging is not a real job where you actually do something to
earn your salary. Begging is just a way to sit around without contributing to society.
11Nr.512Nr.9013Nr.78
42
4. 2 Discussion This chapter is mainly devoted to a discussion of the results and interpretation of them
in the light of previous research. The start of the chapter deals with the general results,
their implications and their relationship to earlier research. The second part discusses
the difference in results by party page and how we can interpret this. The third part
discusses the elements of dehumanization found in this study in relation to other studies,
while the final part assesses the applicability of our results to other issues and areas of
research.
4.2.1. Making Sense of the 22 percent The degree to which dehumanization was used in the discussions was 22%. Even
though the results varied depending on party page, dehumanization was to some extent
present in all cases. Which means that more than a fifth of the discussants consciously
or subconsciously believe that people who beg are less than human. That beggars lack
some qualities that they or their kin intrinsically possess. As previous research shows
that dehumanization reduces the propensity to help (Andrighetto 2015; Cuddy 2007),
there is reason to believe that beggars are less likely to be extended a helping hand than
others. As dehumanization is often used as a justification for differential treatment
(Volpato & Andrighetto 2015), it is easier to infringe on the rights of people who are
victims of dehumanization. It is thus critical to investigate the possible consequences
for democracy in terms of equal participation and voice. Beyond the potential political
consequences, the feeling of being dehumanized often manifests itself in internalized
dehumanization. Victims of dehumanization experience guilt, shame and perceive
themselves as having lower social status (Hartley & Fleay 2017; Bastian & Haslam
2011; Volpato & Contarello 1999).
Dehumanization is being used almost exclusively as a way to express negative
sentiments towards beggars. Even though only a few people explicitly used
dehumanizing statements to argue for a ban on begging or deportation of beggars, the
results clearly show that they are being used to express negative attitudes towards
beggars in the debate. By targeting and excluding certain societal groups, these results
have similarities with right wing-populist rhetoric (Mudde 2007; Wirz et al. 2018).
Populist style rhetoric often assigns blame and depicts subjects and people in a
43
polarizing manner. This blame is often assigned to immigrants and asylum seekers (Bos
et al. 2013; Engesser 2017; Jagers & Walgrave 2007).
Another reason to further investigate the use of discursive dehumanization in rhetoric is
that social psychologists have found a correlation between dehumanizing depictions of
immigrants and negative attitudes towards immigration (Utych 2018; Goméz-Martínez
& de la Villa Moral- Jiménez 2018). Conclusions from these studies suggest that the use
of dehumanizing imagery have repercussions beyond the person who dehumanizes. If
this applies to the investigation at hand, a lot of people run the risk being influenced by
these views. Especially since these are public discussions on popular pages which
everyone has access to.
4.2.2 Dehumanization and Party support The results established that dehumanization is used in the debate. However, the results
also showed a discrepancy in the number of dehumanizing comments depending on the
party page. Even though we cannot know the extent to which people who interact with
party pages are also supporters of that party, we can at least conclude that there is a
discrepancy. It is possible to, for example, conduct a survey on dehumanization where
participants answer questions on party preference, political identity, and ideology to be
able to draw such conclusions.
In this case, the difference between the parties engaged in the debate could be a result of
a difference in ideology and party support, but it can also be a result of party page
moderation. Unfortunately, we cannot know to what extent the party pages are
moderated (Falasca, Dymek and Grandien 2017). Both the Social Democrats and the
Moderate Party reserves the right to remove hateful and offensive comments
(Socialdemokraterna 2018; Moderaterna 2018). The Sweden Democrats do not state
that they will remove hateful comments, but encourage the use of respectful language.
Despite this, a large proportion of the material in this study violates the terms and
conditions for commenting on these pages. Which suggests that the pages are not
completely moderated.
44
The high rate of dehumanization on the Sweden Democrat Party page could also be a
result of comments by people placed further to the right than SD. Still, previous
research finds that people tend to interact with others who share their political
convictions (Jungerr 2016; Internetstiftelsen 2018). Which suggests that people might
seek out their kin by interacting with their own party, and people who support the same
party as themselves. Although many discussants used the party forums to express their
support or dissatisfaction with the party proposals, and support for the parties
themselves, we cannot equate a comment with actual support. The results are thus
inconclusive in this regard.
What we can draw from the discrepancy between different parties is that the rate of
dehumanization can differ depending on the forum. The general social media
demographic is more liberal, more educated and more positive to immigration than the
rest of the population (Mellon & Prosser 2017; Internetstiftelsen 2018; Sandberg &
Bjereld 2015). At the same time, hate speech and antagonistic comments flourish on the
Internet (Iselidh 2018; Näthatsgranskaren 2018). If both of these camps are right, it
suggests that social media users are liberal on a general level, while isolated extreme
forums also exist. The results of this study are placed somewhere in between.
4.2.3 The Relative Prevalence of Moral Dehumanization As previously illustrated, the results show a dominance of mainly two elements of
dehumanization: immorality with 56% and the combined category uncultured/unrefined
with 34% (Figure 3). The implication of combining the category is of course, that the
relative frequency increases. However, these categories separately would still
overshadow the results of the remaining elements. Besides, there were few
unambiguous cases between the two categories, and the analytical tool needs further
development if they are to be applied separately.
The fact that immorality and lack of culture/coarseness are the most frequently
occurring elements probably says more about the debate in particular and debates on
immigration issues in general, than it does about dehumanization as a phenomenon.
Previous research shows that different groups are dehumanized in different ways
(Haslam 2006). The results of this study are consistent with previous research on
45
dehumanization of immigrants, which highlight moral dehumanization of refugees as
especially common (Esses et al. 2008). Even though these groups and the related
political issues are different, dehumanization of immigrants in general, have been
known to contain moral elements (Haslam 2006).
By utilizing cultural stereotypes, it is easy to use dehumanization as an argument for
why a particular group does not belong. So, it is natural that dehumanizing arguments
hinges on traditional stereotypes to gain credibility. The findings here confirm previous
studies on negative stereotypes of Roma people (Orosz et al. 2018), which go under the
umbrella of subtle dehumanization. Among people proposing a ban, a majority of
arguments were built on traditional stereotypes of Roma people as idle, filthy, with low
moral and hostile behavior. Cultural stereotypes might also be the reason why some
elements of animalistic dehumanization are not present in this study, such as
irrationality and maturity. Earlier studies of dehumanization have shown that people
with intellectual disabilities and African people are the most usual targets of this type of
dehumanization (O'Brien 2003b; Haslam 2006). In this paper, I investigated accounts of
animalistic dehumanization, which means that I have not examined mechanistic
dehumanization, and the deprivation of human qualities such as empathy, agency,
individuality etc. (Haslam 2006). As explained in the methods chapter, I deselected
these forms because of feasibility, objectivity, and findings by previous research. Since
they were not included as indicators, we can only rely on previous research to deduce
whether they are present in the debate or not.
4.2.4 The Applicability of the Results This study has consequently found that dehumanization is being used in the debate on
street begging in everyday politics. As previously explained, it is difficult to estimate
population size, and acquire representative samples through random sampling. For our
results, this means that we cannot claim that these figures are a perfect representation of
the entire debate. Given that the material in this paper comes from popular public
forums, we can expect the results to differ or be higher in less public discussion forums
online. The majority of our results (94%) consist of indirectly and partly dehumanizing
comments. Only 6% were explicit and absolute (Figure 3). Considering that blatant
dehumanization correlates with signs of aggression and support for aggressive behavior
46
(Bruneau, Kteily & Laustsen 2018), we should not take lightly on this figure. Given the
risk of being prosecuted for posting blatantly dehumanizing comments and the right to
remove these comments from platforms such as Facebook, it is unlikely that we would
find a higher rate of this type of comments on official channels without anonymity.
Even though the prospects to make inferences from the results of a case study are low
(Gerring 2013), we can at least estimate whether 22% is low or high in relation to other
political topics. Previous research gave some indications to whether we would be able
to find examples of dehumanization, without appreciating the expected share of
dehumanization. Studies within social psychology find that dehumanizing views are
held against outgroups even in democratic societies. As several studies also show that
dehumanization can be transformed into action (Bruneau, Kteily & Laustsen 2018;
Andrighetto 2015; Cuddy 2007), finding at least some evidence of dehumanization was
anticipated in this study. Given that subtle dehumanization in discourse has not been
previously studied, there was no indication of the specific amount.
As explained in the theory chapter, dehumanization is often used as a justification of
violence or aggression, distancing individuals from an event that might invoke feelings
of guilt or compassion, or legitimizing the current social position of a group (Volpato
& Andrighetto 2015). In this particular political issue, all three are possible reasons
why people dehumanize beggars. In this case, dehumanization can work as a way to
justify violence or aggression on the side of the police. Many people argue for the
removal of beggars from the streets in particular and the country in general.
Dehumanization as a way of distancing oneself from feelings of guilt or compassion is
probably especially relevant in this case as we are constantly reminded and exposed to
the misery and poor conditions of beggars, unlike many other groups. In order to avoid
feelings of responsibility or guilt when passing a beggar outside a grocery store, it is
probably easier to dehumanize them and continue with our lives as usual. From our
results, we can observe that many dehumanizing arguments were based on anecdotal
evidence, true or not. Which suggests that many people experience seeing beggars
everywhere, and must find some way to deal with this. Dehumanization can also serve
as an explanation to why you are not the one sitting outside on a cold winter day asking
strangers for money. The idea that some people are less than human because they are
immoral, lack sophistication or intelligence makes it easy to justify that they do not
47
deserve to have the same living standards as yourself. Taken together, this suggests that
the results in this regard are not applicable to other cases or political issues where
outgroups are dehumanized. Furthermore, this indicates that the results of this study
represent a maximum level, and we should probably expect the number of
dehumanizing comments to be lower in debates on less contentious issues in Sweden
where people are not exposed to the issue in their daily lives.
48
5 Conclusion This study set out to answer whether and how dehumanization is used in the debate on
street begging in everyday politics. This concluding chapter begins by answering the
question of whether dehumanization is used. The second part gives an answer to how
dehumanization is used and reviews the analytical tool. The last section contains
implications as well as the main theoretical, empirical and methodological conclusions
for further research.
5.1 Whether Dehumanization is Used The short answer to the question of whether dehumanization is used is yes. It is being
used, at the rate of 22%, and predominantly to express hostility and negative sentiments
towards beggars. What the results mean in general is that discursive dehumanization is
part of the debate in everyday politics, and not just a propaganda tool of in war and
genocide. This study corroborates findings from previous studies, which have found a
connection between dehumanizing attitudes and negative attitudes towards immigration
(Utych 2018; Goméz-Martínez & de la Villa Moral- Jiménez 2018). In a broader sense,
this study contributes to the general dehumanization discourse by showing that people
are also willing to express their dehumanizing views openly. In light of these results, it
is key to investigate whether dehumanization can lead to restrictions of rights and
liberties in democratic societies. Avenues for future research can be whether
dehumanization is intently used, or whether it is explicitly being used to support
arguments for restricting immigration.
To nuance the answer to whether dehumanization is used, our results also showed that
the extent largely varies depending on the party page. We can conclude that there was a
difference in the rate of dehumanizing comments between party pages. These results
can provide a foundation for further research on dehumanization in political science.
This question could be investigated through survey research connecting political
affiliation and ideology to dehumanization. Given that threat perception of immigrants
exist, and that threat is an antecedent to dehumanization, a possible venue for future
studies could thus be to examine the relationship between the perceived threat from
49
immigration, dehumanization and political attitudes among supporters of radical right-
wing populist parties.
5.2 How Dehumanization is Used The answer to the question of how dehumanization is used is mainly by claiming that
they are immoral, uncivilized, and unrefined. The results of this study show that
dehumanization functions as a way to negatively depict beggars. In some cases,
dehumanization functions as an argument for banning or removing beggars. In this
particular issue, dehumanization mainly manifests itself as the deprivation of someone’s
morality, or civility and refinement. The results from this study show that 56% of the
dehumanizing comments have moral elements. As these results are consistent with
research on dehumanization of refugees in for example Canada (Esses et al. 2008), they
can lay the foundation for new testable assumptions. One such avenue is to test whether
moral dehumanization is central to dehumanization of immigrants in general.
The analytical tool created for the task of finding discursive dehumanization is one of
the main contributions of this paper. With some modification, it can easily be applied to
other European countries, political spheres or issues. Further applications would not
only provide ground for cross-country comparisons, but it would also benefit the
development of the tool. Given that Sweden is considered to be the last bastion of
liberal values, we can expect to find evidence of dehumanizing discourse in other
countries as well (Embury-Dennis 2018; Stavrou 2018; Al Jazeera News 2018).
Overall, the analytical tool was successfully applied as an initial test. However, further
development of the tool is necessary. If the categories overlap, it is not fruitful to
distinguish between different elements of dehumanization. Since the aggregated
category is rather broad and diverse, it would be wise to construct lack of culture and
refinement as clear and separate elements, in the same way as morality, rationality, and
maturity are differentiated. Constructing a scale with degrees of discursive
dehumanization can further develop the analytical tool. This would provide more
nuances to the results than the dichotomous variables that are used in this study can.
50
5.3 The Future of Dehumanization in Political Science
There is no consensus among political science scholars as to whether the Internet in
general and social media, in particular, enhances democracy, or fosters extremism.
Facebook and Twitter are open to everyone, and one of the more accessible means of
politic participation, and the general social media user is more liberal and more
educated than the general population (Internetstiftelsen 2018; Mellon & Prosser 2017;
Margetts 2019). Yet, these platforms are used as means for extremists to mobilize and
shape their political identity (Caiani & Parenti 2013; Awan 2017). This study paints a
dark image of social media. However, unlike Caiani and Parenti (2013), and Awan
(2017), this study does not investigate far-right forums, or involve extremists. It
involves politically engaged citizens, of which, many express opinions in their own
name. Since at least a fifth of the comments is dehumanizing in a subtle way, this study
places itself in the grey area between democracy and extremism. Scholars who study
new forms of democratic participation need to embrace this darker side of social media
as well. For political science to understand the relationship between social media and
democracy, we need to study it as more than a means for social change, or a way of
enhancing democratic values. If we see that democratic governance is more than just a
top-down perspective, where citizens themselves can also steer the political direction, or
influence the political agenda, we need to pay attention to how citizens discuss political
issues and express their political attitudes.
As dehumanization is conceptually different from other phenomena and can have such
dire consequences, it is vital to further the study of how dehumanization is used and
where it comes into play. These results can provide a foundation for studies of
dehumanization in political science and the understanding of the political impact of
dehumanization in democracies. Historically, women and people with disabilities have
been denied the right to vote, as they allegedly lack rationality. If we do not believe that
someone is our equal, then why should we treat them like such? The results of this
study found that many people try to find arguments to support that there is a difference
between beggars and beggars. That Swedish charity organizations and churches should
not be banned from collecting money in the streets even though they engage in the same
activity.
51
If dehumanization and extreme forms of prejudice are a part of public opinion, we need
to recognize it. Regardless of whether these attitudes stem from political rhetoric, or
whether politicians take advantage of already existing attitudes (Berning & Schlueter
2015), they can have a serious political impact. Large-N studies show that
psychological factors such as threat perception are important factors behind anti-
immigrant sentiments (Norris & Ingelhart 2016; Lucassen & Lubbers 2012). At the
same time, dehumanization is a mediator between sense of threat and negative attitudes
towards immigration (Maoz & MacCauley 2008; Louis, Esses & Lalonde 2013; Prati et
al. 2016). As previously written, 96-98% of Sweden Democratic voters believe that
immigration leads to increased criminality, 80% of the Moderate Party and 43% of the
Social Democrats share this view (Jylhä, Rydgren & Strimling 2018). The results of this
study showed that all party pages have comments, which include moral dehumanization
of beggars. These results combined suggest that future research should investigate this
possible connection between threat perception, dehumanization and anti-immigration
attitudes.
Today, everyday politics mainly takes place on social media and a fifth of the
population in Sweden state that they express their political opinions online (Highfield
2016; Internetstiftelsen 2018). For democracy researchers, social media networks
provide a unique insight into the natural behavior and beliefs of citizens (Moreno et al.
2013). Signing an online petition or tweeting a political statement is far more accessible
and resource effective than joining a party or going door-to-door (Margetts 2019). The
results of this study shows, not only that dehumanizing language is expressed online,
but also that many citizens engage in discussions on political issues on heated topics
online. This suggests that political science should pay more attention to the political
activities of citizens on social media.
Other researchers have found that it is difficult to gather comprehensive data on one
topic on Twitter (Lorentzen & Nolin 2017). In accordance, this study was able to map
inputs into the debate, and only fragments of Twitter-discussions. Facebook, on the
other hand, enables researchers to follow entire discussions chronologically. Another
related methodological lesson to take home from this study is that the interaction
between the people and political parties in Sweden is easier to follow on Facebook. As a
consequence, if discussion as a form is important, Facebook is the better option.
52
6 List of References Aftonbladet. (2014). Därför stänger vi våra kommentarsfält. Aftonbladet. June 16.
https://www.aftonbladet.se/nyheter/a/ddGmGw/darfor-har-vi-stangt-vara-kommentarsfalt Retrieved [2018-10-25]
Al Jazeera News. (2018). Refugees, crime, environment in spotlight as Swedes head to
polls. Al Jazeera. September 9. https://www.aljazeera.com/news/2018/09/refugees-crime-environment-spotlight-swedes-head-polls-180905095321894.html Retrieved [2018-10-03]
Amoral. (n.d.). Oxford online dictionary.
https://en.oxforddictionaries.com/definition/amoral Retrieved [2018-10-04]
Anderson, B. (2017). The Politics of Pests: Immigration and the Invasive Other. Social
Research, Vol. 84:1. Andrighetto, L., Baldissarri, C., Lattanzio, S., Loughnan, S., & Volpato, C. (2014).
Humanitarian aid? Two forms of dehumanization and willingness to help after natural disasters. British Journal of Social Psychology, Vol. 53:3, pp. 573–584.
Awan, I. (2017). Cyber-extremism: Isis and the power of social media. Society,
Vol. 54:2, pp. 138-149.
Bastian, B., Haslam N. (2011). Experiencing dehumanization: cognitive and emotional effects of everyday dehumanization. Basic Applied Social Psychology, Vol. 33, pp. 295–303.
Bastian, B., Laham, S. M., Wilson, S., Haslam N. and Koval, P. (2011). Blaming, praising, and protecting our humanity: The implications of everyday dehumanization for judgments of moral status. British Journal of Social Psychology, Vol. 50, pp. 469-483.
Bellin, E. (2012). Reconsidering the Robustness of Authoritarianism in the Middle East: Lessons from the Arab Spring. Comparative Politics, Vol. 44:2, 127-149.
Benkler, Y., Faris, R. and Roberts, H. (2018). Network Propaganda: Manipulation, Disinformation, and Radicalization in American Politics [Ebook]. Available from: Oxford Scholarship Online
Bleiker, R., Campbell, D., Hutchison, E. & Nicholson, X. (2013). The visual dehumanisation of refugees, Australian Journal of Political Science, 48:4, pp. 398-416.
Bobba, G., Cremonesi, C., Mancosu, M., & Seddone, A. (2018). Populism and the Gender Gap: Comparing Digital Engagement with Populist and Non-populist
53
Facebook Pages in France, Italy, and Spain. The International Journal of Press/Politics, Vol. 23:4, pp. 458–475.
Bos, L., Van, D.B. and De Vreese, C.,H. (2013). An experimental test of the impact of
style and rhetoric on the perception of right-wing populist and mainstream party leaders. Acta Politica, Vol. 48, no. 2, pp. 192-208.
Bruneau, E., Kteily, N., and Laustsen, L. (2018). The unique effects of blatant
dehumanization on attitudes and behavior towards Muslim refugees during the European ‘refugee crisis’ across four countries. European Journal of Social Psychology.
Caiani, M. & Parenti, L. (2013). European and American Extreme Right Groups and
the Internet [Ebook]. Routledge, Abingdon. Available from: ProQuest Ebook Central.
Ceron, A., Curini, L., Iacus, S. M., and Porro, G. (2014). Every tweet counts? How sentiment analysis of social media can improve our knowledge of citizens’ political preferences with an application to Italy and France. New Media & Society, Vol. 16:2, pp. 340–358.
Charteris-Black, J. (2006). Britain as a container: immigration metaphors in the 2005 election campaign. Discourse and Society, Vol. 17:5, pp. 563 – 581.
Coarseness. (n.d.). Cambridge online dictionary. https://dictionary.cambridge.org/dictionary/english/coarseness Retrieved [2018-10-05]
Cuddy, A. J., Rock, M. S., & Norton, M. I. (2007). Aid in the aftermath of Hurricane Katrina: Inferences of secondary emotions and intergroup helping. Group Processes & Intergroup Relations, Vol. 10:1, pp. 107–118.
Dalsklev, M. and Kunst, J. R. (2015). The effect of disgust-eliciting media portrayals on outgroup dehumanization and support of deportation in a Norwegian sample. International Journal of Intercultural Relations, Vol. 47 pp. 28-40.
Davidsson, Palm & Mandre. (2018). Svenskarna och Internet 2018. Internetstiftelsen, pp. 1-121.
DeBono, D. (2013). ‘Less than human’: the detention of irregular immigrants in Malta. Race and Class, Vol. 55:2, pp. 60-81.
DiGrazia, J., McKelvey, K., Bollen, J., Rojas, F. (2013). More Tweets, More Votes: Social Media as a Quantitative Indicator of Political Behavior. PLoS ONE, Vol. 8:11.
Domonkos, S. (2015). The Imitated Public Sphere. In P.A. Simpson and H. Druxes (eds), Digital Media Strategies of the Far Right in Europe and the United States, Lexington Books, Lanham, MD. Available from: ProQuest Ebook Central. [29 October 2018].
54
Dredge, S. (2015). Facebook clarifies policy on nudity, hate speech and other community standards. The Guardian. March 16. https://www.theguardian.com/technology/2015/mar/16/facebook-policy-nudity-hate-speech-standards Retrieved [2018-10-30]
Embury- Dennis, T. (2018). Sweden Democrats: Far-right party claims victory despite
third-place finish in general election. The Independent. September 10. https://www.independent.co.uk/news/world/europe/sweden-democrats-election-exit-poll-latest-social-democratic-party-far-right-moderates-a8530581.html Retrieved [2018-10-03]
Engesser, S., Ernst, N., Esser, F. and Büchel, F. (2017). Populism and social media: how politicians spread a fragmented ideology, Information, Communication & Society, Vol. 20:8, pp. 1109-1126.
Enli, G. (2017). Twitter as arena for the authentic outsider: exploring the social media campaigns of Trump and Clinton in the 2016 US presidential election. European Journal of Communication, Vol. 32:1, pp. 50–61.
Ernst, N., Engesser, S. and Esser, F. (2017). Bipolar Populism? The Use of Anti-Elitism and People-Centrism by Swiss Parties on Social Media. Swiss Political Science Review Vol. 23:3, pp. 253–261.
Esses, V.M., Medianu, S. and Lawson A. S. (2013) Uncertainty, Threat, and the Role of the Media in Promoting the Dehumanization of Immigrants and Refugees. Journal of Social Issues, Vol 69:3 pp. 518-536.
Esses, V.M., Veenvliet, S., Hodson, G. & Mihic, L. (2008). Justice, Morality, and the Dehumanization of Refugees. Social Justice Research, Vol. 21:1, pp. 4-25.
Falasca, K., Dymek, M. and Grandien, C. (2017). Social media election campaigning: who is working for whom? A conceptual exploration of digital political labour. Contemporary Social Science, pp. 1-13.
Gadde, V. and Harvey, D. (2018). Creating New Policies Together. Twitter Blog. September 25. https://blog.twitter.com/official/en_us/topics/company/2018/Creating-new-policies-together.html Retrieved [2018-10-25]
Gerring, J. (2011[2009]). The Case Study: What it is and What it Does. In Goodin, Robert E. (red.) The Oxford handbook of political science [Ebook]. Oxford: Oxford University Press.
Gert, Bernard (2004). Common Morality Deciding What to Do [Ebook]. New York: Oxford University Press
Gómez-Martínez, C. and de la Villa Moral-Jiménez, M. (2018) Dehumanization and Islamophobia: attitudes towards the Syrian refugee crisis / Deshumanización e
55
islamofobia: actitudes ante la crisis de los refugiados sirios, Revista de Psicología Social, Vol. 33:2, pp. 215-239.
Granström, K. (2016). Allt fler mediehus stänger sina kommentarsfält. Journalisten.
April 28. https://www.journalisten.se/nyheter/allt-fler-mediehus-stanger-sina-kommentarsfalt Retrieved [2018-10-25] Gray, H. M., Gray, K., & Wegner, D. M. (2007). Dimensions of mind perception. Science, 315: 5812, pp. 619.
Haagensen, L. and Croes, M. (2012). Thy Brother’s Keeper?: The Relationship between Social Distance and Intensity of Dehumanization during Genocide. Genocide Studies and Prevention, Volume 7: 2-3, pp. 223-250.
Hagan, J. and Rymond-Richmond, W. (2008). The Collective Dynamics of Racial Dehumanization and Genocidal Victimization in Darfur. American Sociological Review, Vol. 73: 6, pp. 875-902.
Halpern, D. and Gibbs, J. (2013). Social media as a catalyst for online deliberation? Exploring the affordances of Facebook and YouTube for political expression. Computers in Human Behavior, Vol. 29 pp.1159–1168.
Harlow, S. and Johnson T. (2011). Overthrowing the Protest Paradigm? How The New York Times, Global Voices and Twitter Covered the Egyptian Revolution. International Journal of Communication, Vol. 5, pp. 1359-1374.
Harris, L. T., and Fiske, S. T. (2011). Perceiving humanity or not: A social neuroscience approach to dehumanized perception. In A. B. Todorov, S. T. Fiske, & D. A. Prentice (Eds.), Social neuroscience: Toward understanding the underpinnings of the social mind, New York, NY: Oxford University Press, 1st ed., pp. 123–134.
Hartley, L. and Fleay C. (2017). “We are Like Animals”: Negotiating Dehumanising Experiences of Asylum-Seeker Policies in the Australian Community. Refugee Survey Quarterly, Vol. 36, pp. 45–63.
Haslam, N. (2006). Dehumanization: An Integrative Review. Personality and Social Psychology Review, Vol. 10: 3 pp. 252–264.
Haslam, N., Kashima, Y., Loughnan, S., Shi, J., & Suitner, C. (2008a). Subhuman, inhuman, and subhuman: contrasting humans with nonhumans in three cultures. Social Cognition, 26:2, pp. 248-258.
Haslam N. and Loughnan S. (2014). Dehumanization and infrahumanization. Annual Review of Psychology, Vol. 65, pp. 399-423.
Haslam, N., Loughnan, S., Kashima, Y. and Bain, P. (2008b). Attributing and denying humanness to others, European Review of Social Psychology, Vol. 19:1, pp. 55-85.
56
Haslam, N. and Stratemeyer, M. (2016). Recent research on dehumanization. Current Opinion in Psychology Vol. 11, pp. 25-29.
Haque O. and Waytz A. (2012). Dehumanization in medicine: causes, solutions, and functions. Perspectives on Psychological Science, Vol. 7, pp. 176–86.
Highfield, Tim. (2016). Social media and everyday politics. Cambridge: Polity.
Hodson, G. and MacInnis, C. C. (2012). Intergroup bias toward “Group X”: Evidence of prejudice, dehumanization, avoidance, and discrimination against asexuals. Group Processes & Intergroup Relations, Vol.15:6, pp. 725–743.
Hooghe, M., and Dassonneville, R. (2018). Explaining the trump vote: The effect of racist resentment and anti-immigrant sentiments. PS, Political Science & Politics, Vol. 51:3, pp. 528-534.
Immaturity. (n.d.). Cambridge online dictionary. https://dictionary.cambridge.org/dictionary/english/immature?q=immaturity Retrieved [2018-10-05]
Inglehart, R. F. and Norris, P. (2016). Trump, Brexit, and the Rise of Populism: Economic Have-Nots and Cultural Backlash. HKS Working Paper No. RWP16-026.
Internetstiftelsen. (2018). Svenskarna och Internet –Valspecial 2018. Svenska Internetstiftelsen
Irrationality. (n.d.). Oxford online dictionary. https://en.oxforddictionaries.com/definition/irrationality Retrieved [2018-10-03]
Iselidh, A. (2018). 70-åring jämförde invandare med djur – åtalas. SVT Nyheter. November 21. https://www.svt.se/nyheter/lokalt/uppsala/70-aring-skrev-nedsattande-om-araber-jamforde-med-djur Retrieved [2018-11-22]
Ivarsflaten, E. (2008). What unites right-wing populists in Western Europe? Re-examining grievance mobilization models in seven successful cases. Comparative Political Studies, Vol. 41:1, pp. 3–23.
Jagers, J., and Walgrave, S. (2007). Populism as political communication style: An empirical study of political parties’ discourse in Belgium. European Journal of Political Research, Vol. 46:3, pp. 319–345.
Jardina, A., McElwee, S. and Piston, S. (2016). How do Trump Supporters See Black People? Slate. November 7. http://www.slate.com/articles/news_and_politics/politics/2016/11/the_majority_of_trump_supporters_surveyed_described_black_people_as_less.html Retrieved [2018-10-11]
57
Jungherr, A. (2016) Twitter use in election campaigns: A systematic literature review, Journal of Information Technology & Politics, Vol. 13:1, pp. 72-91.
Jylhä, K., Rydgren, J. and Strimling, P. (2018). Sverigedemokraternas väljare: Vilka är de, var kommer de ifrån och vart är de på väg? Institutet för Framtidsstudier, Forskningsrapport 2018:2, pp. 1-100.
Kelman, H. C. (1973). Violence without moral restraint: Reflections on the dehumanization of victims and victimizers. Journal of Social Issues, Vol. 29:4, pp. 25–61.
Kteily, N., Bruneau, E., Waytz, A., & Cotterill, S. (2015). The ascent of man: Theoretical and empirical evidence for blatant dehumanization. Journal of Personality and Social Psychology, Vol. 109:5, pp. 901-931.
Larsson, A.O. and Moe, H. (2012). Studying political microblogging: Twitter users in the 2010 Swedish election campaign. New Media & Society, Vol. 14:5, pp. 729–747.
Leyens, J., Paladino, P.M., Rodriguez-Torres, R., Vaes J., Demoulin, S., Rodriguez, A., Gaunt, R. (2000). The emotional side of prejudice: The attribution of secondary emotions to ingroups and outgroups. Personality and Social Psychology Review, Vol. 4:2, pp. 186-197.
Lindén, M., Björklund, F. and Bäckström, M. (2015). What makes authoritarian and socially dominant people more positive to using torture in the war on terrorism? Personality and Individual Differences, Vol. 91, pp. 98–101.
Lorentzen, D. G., & Nolin, J. (2017). Approaching Completeness: Capturing a Hashtagged Twitter Conversation and Its Follow-On Conversation. Social Science Computer Review, Vol. 35:2, pp. 277–286.
Louis, W. R., Esses, V. and Lalonde, R. N. (2013). National identification, perceived threat, and dehumanization as antecedents of negative attitudes toward immigrants in Australia and Canada. Journal of Applied Social Psychology, Vol. 43, pp. 156-165.
Lucassen, G. and Lubbers, M. (2012) Who Fears What? Explaining Far-Right-Wing Preference in Europe by Distinguishing Perceived Cultural and Economic Ethnic Threats. Comparative Political Studies, Vol. 45:5, pp. 547–574.
Luft, A. (2015). Toward a Dynamic Theory of Action at the Micro Level of Genocide: Killing, Desistance, and Saving in 1994 Rwanda. Sociological Theory. Vol. 33:2, pp. 148-172.
Maoz, I. and McCauley, C. (2008). Threat, dehumanization, and support for retaliatory aggressive policies in asymmetric conflict. Journal of Conflict Resolution, Vol. 52, pp. 93–116.
Margetts, H. (2019). Rethinking Democracy with Social Media. The Political Quarterly, Vol. 90, pp. 107-123.
58
Mellon, J. and Prosser, C. (2017). ‘Twitter and Facebook are not representative of the
general population: Political attitudes and demographics of British social media users’, Research & Politics. Vol. 4:3, pp. 1-9.
Miles, T. (2018). U.N. investigators cite Facebook role in Myanmar crisis. Reuters. March 12. https://uk.reuters.com/article/us-myanmar-rohingya-facebook/u-n-investigators-cite-facebook-role-in-myanmar-crisis-idUKKCN1GO2PN Retrieved [2018-10-31]
Moreno, M. A., Goniu, N., Moreno, P.S. and Diekema, D. (2013). Ethics of Social Media Research: Common Concerns and Practical Considerations. Cyberpsychology, Behaviour and Social Networking, Vol. 16:9, pp. 708–713.
Mudde, C. (2007). Populist radical right parties in Europe. [Ebook] Cambridge: Cambridge University Press.
Norris, P. (2005). Radical right: Voters and parties in the electoral market. Cambridge, UK: Cambridge University Press.
Nya Moderaterna. (2018). Riktlinjer för kommentarer. Facebook @Moderaterna. May 16. Retrieved [2018-12-04]
Näthatsgranskaren. (2018). Om oss. Näthatsgranskaren Retrieved [2018-12-11] http://xn--nthatsgranskaren-vnb.se/
Obtrusive. (n.d.). Merriam-Webster online dictionary. https://www.merriam-webster.com/dictionary/obtrusive Retrieved [2018-12-02]
O'Brien, G. V. (2003a). Indigestible Food, Conquering Hordes, and Waste Materials: Metaphors of Immigrants and the Early Immigration Restriction Debate in the United States. Metaphor and Symbol, Vol. 18:1, pp. 33-47.
O'Brien, G. V. (2003b). People with cognitive disabilities: The argument from marginal cases and social work ethics. Social Work, Vol. 48:3, pp. 331-7.
Orosz, G., Bruneau, E., Tropp, L. R., Sebestyén, N., Tóth‐Király, I. and Bőthe, B. (2018). What predicts anti‐Roma prejudice? Qualitative and quantitative analysis of everyday sentiments about the Roma. Journal of Applied Social Psychology, Vol. 48, pp. 317–328.
Placek, M. A. (2017). #Democracy: social media use and democratic legitimacy in Central and Eastern Europe. Democratization, Vol. 24:4, pp. 632-650.
59
Prati, F., Crisp, R. J., Meleady R. and Rubini, M. (2016). Outgroups Through Multiple Categorization: The Roles of Individuation and Threat. Personality and Social Psychology Bulletin, Vol. 42:4, pp. 526 – 539.
Roscoe, P. (2007). Intelligence, Coalitional Killing, and the Antecedents of War. American Anthropologist, Vol. 109: 3, pp. 485-495.
Rydgren, J. (2007). The sociology of the radical right. Annual Review of Sociology, Vol. 33, pp. 241–262.
Rydgren, J. (2008). Immigration sceptics, xenophobes or racists? Radical right-wing voting in six West European countries. European Journal of Political Research, Vol. 47, pp. 737-765.
Sandberg, L. and Bjereld, U. (2015). Politiskt deltagande och engagemang i en digital kontext i Annika Bergström, Bengt Johansson, Henrik Oscarsson & Maria Oskarson (ed.) Fragment. Göteborgs universitet: SOM-institutet.
Santa Ana, Otto. (2002). Brown Tide Rising: Metaphors of Latinos in Contemporary American Public Discourse [Ebook]. Austin: University of Texas Press, Project Muse
Santa Ana, Otto. (1999). ‘Like an Animal I Was Treated’: Anti-immigrant Metaphor in US Public Discourse. Discourse & Society 10: 2: 191–224.
Savage R. 2013. Modern genocidal dehumanization: a new model. Patterns of Prejudice 47: 139–61.
Self-restraint. (n.d.). Oxford online dictionary. https://en.oxforddictionaries.com/definition/self-restraint Retrieved [2018-10-03]
Schlozman, K. L., Verba, S., & Brady, H. E. (2010). Weapon of the strong? Participatory inequality and the Internet. Perspectives on Politics, Vol. 8:2, pp. 487–509.
Smith, David Livingstone (2012[2011]). Less than human: why we demean, enslave, and exterminate others. New York: St. Martin's.
Socialdemokraterna. (2018). Vår policy för bra samtalsklimat. Facebook @Socialdemokraterna. July 5. Retrieved [2018-12-04]
Spierings, N., & Jacobs, K. (2014). Getting personal? The impact of social media on preferential voting. Political Behavior, Vol. 36:1, pp. 215-234.
Stavrou, D. (2018). Sweden Is Europe's Last Liberal Bastion. The Elections Might Change That. Haaretz. September 9. https://www.haaretz.com/opinion/.premium-europe-s-last-liberal-bastion-1.6459584 Retrieved [2018-10-03]
60
Stephan, W. G., and Stephan, C. W. (2000). An integrated theory of prejudice. In S.
Oskamp (Ed.), Reducing prejudice and discrimination: The Claremont Symposium on Applied Social Psychology. Mahwah, NJ: Lawrence Erlbaum, pp. 23-45.
Stephan, W. G., Ybarra, O., Martnez, C. M., Schwarzwald, J., & Tur-Kaspa, M. (1998). Prejudice toward immigrants to Spain and Israel: An integrated threat theory analysis. Journal of Cross-Cultural Psychology, Vol. 29, pp. 559-576.
Steuter, E. and Wills, D. (2010). ‘The vermin have struck again’: dehumanizing the enemy in post 9/11 media representations. Media, War & Conflict, Vol. 3:2, pp. 152 – 167.
Svenska Dagstidningar. (2018). Kungliga Biblioteket https://tidningar.kb.se/ Retrieved [2018-09-23]
Sverigdemokraterna. (2018). Välkommen att göra din röst hörd. Facebook @Sverigedemokraterna. November 8. Retrieved [2018-12-04]
Swigger, N. (2013). The online citizen: Is social media changing citizens' beliefs about democratic values? Political Behavior, Vol. 35:3, pp. 589-603.
Tipler, C. and Ruscher, J. (2017). Dehumanizing representations of women: the shaping of hostile sexist attitudes through animalistic metaphors. Journal of Gender studies.
Utych, S. M. (2018). How Dehumanization Influences Attitudes toward Immigrants. Political Research Quarterly, Vol. 71:2, pp. 440–452.
Uncultured. (n.d.). Oxford online dictionary. https://en.oxforddictionaries.com/definition/uncultured Retrieved [2018-10-03]
Viki, T., Winchester, L., Titshall, L., Chisango, T., Pina, A. & Russell, R. (2006). Beyond Secondary Emotions: The Infrahumanization of Outgroups Using Human-Related and Animal-Related Words. Social Cognition, Vol. 24:6, pp. 753-775.
Volpato, C. and Andrighetto, L. (2015). Dehumanization. In Wright, J.D. International Encyclopedia of the Social & Behavioral Sciences, Vol. 26 (2nd ed.), Amsterdam: Elsevier, Inc, pp. 31-37.
Volpato, C. and Contarello, A. (1999). Towards a social psychology of extreme situations: Primo Levi's If This is a Man and social identity theory. European Journal of Social Psychology, Vol. 29, pp. 239-258.
Wirz, D. S., Wettstein, M., Schulz, A., Müller, P., Schemer, C., Ernst, N., Esser, F. and Wirth, W. (2018). The Effects of Right-Wing Populist Communication on
61
Emotions and Cognitions toward Immigrants. The International Journal of Press/Politics, Vol. 23:4, pp. 496 – 516.
Wodak, R. (2013). ‘Anything Goes!’ – The Haiderization of Europe. In R. Wodak, M. KhosraviNik & B. Mral (Eds.) Right-Wing Populism in Europe: Politics and Discourse. London: Bloomsbury Academic, pp. 23-38.
Xenos, M., and Moy, P. (2007). “Direct and Differential Effects of the Internet on Political and Civic Engagement.” Journal of Communication, Vol. 57:4, pp. 704–718.
Zhang, W., Johnson, T. J., Seltzer, T., & Bichard, S. L. (2010). The Revolution Will be
Networked: The Influence of Social Networking Sites on Political Attitudes and Behavior. Social Science Computer Review, Vol. 28:1, pp. 75–92.
62
7 Appendix
Appendix 1: Indicators
Qualitybeing
denied
Indicator Examplefromtheliterature Possible
examplesfrom
thedebate
Civility Uncultured,
uncivilized
“Notcharacterizedby
goodtaste,manners,or
education”(OxfordOnline
Dictionary)
”(ofasocietyorcountry)
withoutwhatis
thoughttobe
ahighlydevelopedculture
andwayoflife,or(of
aperson)rudeand
notshowing
careforothers”
(CambridgeOnline
Dictionary)
“Backwardness”(Bastianand
Haslam2011)
”Asignificantportionofthe
Gypsiesareunfitforco-existence,
notfittoliveamonghuman
beings”(Oroszetal.2018)
“Dirty,livinglikeanimals”
(LivingstoneSmith2011)
Evokingdisgustbybeingdirtyand
livinginunsanitaryconditionsbut
alsothroughlackofeducationand
wayofliving(Dalsklev&Kunst
2015).
“Groupxisimpolite”(Haslametal.
2008b)
“theyare
underdeveloped”,
“uneducated”,“not
fittoworkhere”,
“notfitforthis
society”
“Theyaredirty,
filthy”
“Theylivein
dirty/unsanitary
environments”,
“theydefecate
anywhere”
“theyareimpolite”,
“Theyare
ungrateful/neversay
thanks”,“theyare
rude”
Refinement Coarseness:crudeorunrefined”Behaviourorlanguage
thatiscrudeand
offensive”(Cambridge
OnlineDictionary)
Seeingsomeoneas
unsophisticated,withoutlanguage
andtheabilitytocommunicate
withothers(Bastian&Haslam
2011)
“theyare
aggressive”,“they
arevulgar”,“they
areraucous”
Moralsensibility Amorality,lackof
63
self-restraint
“Lackingamoralsense;
unconcernedwiththe
rightnessorwrongnessof
something”(Oxford
OnlineDictionary)
Self-restraint:
”Restraintimposedby
oneselfonone'sown
actions;self-control”
(OxfordOnlineDictionary)
“refugeesarebogusandnotinreal
need”,“refugeesarecheaterswho
trytricktheirwayin”(Essesetal.
2008)
“Immigrantsdon’tfollowtherules”
(Essesetal.2008;Louis,Esses&
Lalonde2013)
“deceptive,cheating,doesnot
obeythelaw,keeptheirpromises
ordotheirduty”(Gert2004).
lackofself-control=someonewho
cannotcontroltheirimpulsesis
unpredictable(Grayetal.2007)
Uninhibited(Oroszetal.2018)
“theytrytotrick
you”,“theymakea
lotofmoney”,“they
arenotinneedor
poor,theyjust
pretendtobe”,
“theyareorganized
criminalnetworks”,
“theyshouldwork
andearntheir
moneylike
everybodyelse”
“theyare
unpredictable”,
“theycreatean
insecure
environment”
Rationality,logic Irrationality,instinct
”Thequalityofbeing
illogicalorunreasonable”
(OxfordOnlineDictionary)
Describingsomeoneas
unintelligent,incompetent(Haslam
&Loughnan2014;Bastian&
Haslam2011)
Drivenbywants,desiresand
emotion,insteadofthinking
(Haslametal.2008b)
“theyarestupid”,
“theyaregullible”,
“theyareeasily
takenadvantageof
becausetheyare
stupid”,“youcannot
explainthingsto
themsotheycan
understand”
Maturity Childlikeness,
immaturity
“Nothavingmuch
experienceofsomething”,
“Notyetcompletely
Seeingsomeoneasimmature,
incompetent(Bastian&Haslam
2011)
“cannottakecareof
themselves”,“are
notfittomaketheir
ownchoices”,“they
can’tmakeiton
theirown”,“they
cannotdoajob
64
grownordeveloped”
(CambridgeOnline
Dictionary)
properly/liveupto
certainstandards”,
“theydon’ttake
responsibilityfor
anything”,“theyare
selfish”(Cambridge
dictionary)
Animals Rats,wolves,vermin,prey(Livingstone
Smith2011;Tipler&Ruscher2014)Naturaldisasteranddiseasemetaphors
Tidalwaves,contagion,virus(Esses,Medianu&Lawson2013;Anderson2017;SantaAna1999;SantaAna2002).
Animalbehaviourandassociatedwords
Swarm,nest,hive,breedingground(Anderson2017;Steuter&Wills2010)
65
Appendix 2: Codesheet Variable 1: Source 0= Facebook 1= Twitter 2= Letter to the editor Variable 2: Gender 0 = Woman 1 = Man 2 = Other/unidentified Variable 3: Tiggeriförbud 0= Against a ban 1= For a ban 2= Unknown Variable 4: Party 0= Unknown 1= Sweden Democrats 2= Socialdemocrats 3= Nya Moderaterna Variable 5: Uncultured 14 0=No 1=Yes Variable 6: Unrefined 0=No 1=Yes 14Variable5and6wereaggregatedafterthecodingprocesswasfinished
Variable 7: Immorality 0=No 1=Yes Variable 8: Irrationality 0=No 1=Yes Variable 9: Immaturity 0=No 1=Yes Variable 10: Animals 0=No 1=Yes Variable 11: Natural disaster 0=No 1=Yes Variable 12: Animal behavior 0=No 1=Yes
Appendix 3: Applying the analytical tool Comment Coding“Outsidethegrocerystoresitsabeggar,alargemanaround40yearsold.Whenpeoplepassbyhemuttersunpleasantthingsandlooksangryandviolentwhenheactslikethecockofthewalk.Thestoregetsalotofcomplaintsbut“can’tdoanything”eventhoughheisintheirprivateland#svpol”(Nr.483)
Variable5:Uncultured1(Yes):“muttersunpleasantthings”:-DemeaningVariable6:Unrefined1(Yes):“looksangryandviolent”-DemeaningVariable10:Animals0(No):“cockofthewalk”isanidiomaticexpression
Translation Coding“ThereisALWAYSwork,beggingispurefuckinglaziness”(Nr.28)
Variable7:Immorality1(Yes):Beggarsareidleanddonotcontributetosociety,inotherwordsbeggarsdonotdotheirduty-Demeaning
Translation Coding“Banbeggingcompletely,[I’m]sotiredofseeingthemateverygrocerystore,bythesubwayetc.”(Nr.13)
Notdehumanizing.Thepersondoesnotstatewhyheorshewantstobanbegging,sowedonotknowwhethertheunderlyingreasonisdehumanizingattitudes.